Slavery in Francophone West Africa

Abstract

Multiple forms of labor coercion were common in West Africa in the first half of the twentieth century. This chapter focuses on two of them: one, based on interior slavery, was rooted in the longue durée of regional history. The other, colonial forced labor, was exogenous and introduced by European colonialists between the mid-nineteenth and the early twentieth century. In Tahoua, the region of today’s Republic of Niger that this chapter focuses on, these two forms of coerced labor co-existed. In 1905 France abolished indigenous slavery in its recently acquired African territories and concurrently introduced a regime of compulsory labor for the building and maintenance of colonial infrastructure. In spite of French legal abolition, indigenous slavery did not die out, but started being contested and resisted more effectively by those who had been enslaved. Those formerly enslaved to local slaveowners and their descendants were also the first to be forcibly recruited to carry out construction work on colonial worksites. These two regimes of coerced labor, with different ideological origins, became intertwined and underwent substantial changes throughout the twentieth century. Their legacies have continued influencing labor relations and forms of personal dependence into the present day.

Introduction¹

Slavery gives slaveholders totalizing power over every aspect of the life of enslaved persons. For this reason, slaves have often been the workers of choice in labor systems that employed disposable workers in order to maximize the slaveholders’ profits. This accounts for the deadly combination of slavery and capitalism, which authors with different political and intellectual agendas generally agree upon.² But we should be weary of generalizations. In the West African Sahel slavery has been both widespread and resilient. And yet this is not because slave labor lent itself to the harshest form of exploitation by profit-maximizing capitalists. Instead, here slavery has been resilient precisely because slavery is more than control over the labor of the enslaved. In the Sahel coercing workers made little economic sense, as there were neither cash crops nor large profits to be made by either private or public employers. Even colonial administrators refrained from committing substantial resources and efforts to the extraction of local labor. This does not mean that they were not violent—they were, in fact, extremely brutal at moments. But punctual explosions of punitive violence cost less than capillary forms of continued policing and control. Colonial violence functioned as a deterrent to revolt in desert-edge territories where the close monitoring of labor was not seen as a reasonable use of resources.

This chapter focuses on Ader, an area roughly overlapping with what today is the administrative Region of Tahoua of the Republic of Niger, located at the Southern edge of the Sahara Desert, where slavery has been peculiarly resilient up until today, when Nigerien abolitionist NGOs are still fighting for its suppression. The French colonial regime abolished slavery in this region in 1905. It introduced forced labor between the 1920s and the 1940s. Wage labor never developed: never has there been a time in the twentieth century when a sizeable proportion—let alone the majority—of Tahoua’s population made a living by earning wages. Since the 1930s, the majority of Tahoua’s poor have been dependent on revenues from seasonal labor migration to earn a living and send remittances to family members who remain in Tahoua’s hinterland villages and struggle with poverty. In general, would-be employers in this region cannot rely on “a combination of favourable soil characteristics and low transport costs […] sufficient to enable them to pay wages and still make a profit.”³ During droughts, the poorest persons of slave descent seek to work for persons of formerly slave-owning classes to be fed. Asked why, in her view, slavery continues today, a woman who had been sold as a child and freed herself a few years ago, answered: “poverty” (hausa: talauci).⁴ With slavery still an option and no paid employment opportunities, poor people must find someone to feed them or die of hunger, someone who’ll agree to act as a paternalist master.⁵

The circumstances of Ader are particular. The rural Sahel is one of the poorest inhabited areas of West Africa. Other, less poor, areas of the sub-region have been studied more extensively. Gareth Austin has discussed cash-crop producing West African regions where after the legal abolition of slavery stable profits allowed employers to pay wages to workers. The latter, in turn, had bargaining power because they knew that their refusal to work would produce significant losses for employers who could not legally force them to work.⁶ In other economically dynamic regions, Frederick Cooper showed that the strikes of African workers galvanized the colonial administration into improving the workers’ conditions.⁷ By contrast, in Ader those who do not migrate have hardly any alternative but to accept the bitter legacies of slavery and appeal to a master’s charity and generosity—attributes that local hegemonic ideologies portray as signs of nobility and piety that will be rewarded in the afterlife.

By comparing slave labor and state-imposed coerced labor in Ader, this chapter shows that in a context where coercion was determined by considerations different from profit maximization, wage labor never developed; forced labor was hard to justify in economic terms and therefore largely ineffective; and slavery remained viable not as a form of extreme labor exploitation in connection to agricultural commercialization, but because it increased safety in the face of environmental adversity. Slavery provided honor, safety, and material advantages to slaveowners and protected enslaved persons from the threat of hunger and extreme destitution. The chapter relies on Marcel van der Linden’s “three moments” framework that dissects coercion into the stages of entry, extraction, and exit. This framework was developed to avoid the biasing influence of the slavery–freedom dichotomy with all the assumptions implicit in the notion of “slavery” as understood and used today.⁸ After summarizing the main processes of labor coercion in twentieth-century West Africa, the chapter compares slave labor and state-imposed forced labor in the Ader region. The conclusion advances some reflections on what this comparison reveals about Sahelian slavery, and how understanding slavery in the Sahel might help us understand slavery as a global-historical phenomenon.

Labor Coercion in Twentieth-Century West Africa

In 1905 France abolished indigenous slavery in its West African territories. More or less simultaneously, it introduced a regime of compulsory labor for the building and maintenance of colonial infrastructure. In spite of French legal abolition, indigenous slavery did not die out in 1905, but merely started being more effectively resisted and contested.⁹ Those formerly enslaved to local slaveowners (and their descendants) were the first to be forcibly recruited to carry out construction work on colonial worksites. In the first half of the twentieth century, colonial administrators controlled the work of statutorily free (non-“slave”) workers. Although the forms and intensity of coercion varied across colonial empires, in the first third of the century forced labor was the most common form of employment of native workers across European colonies. Contracts were exceedingly rare. In 1928 government figures showed that less than 1 percent of workers in French West Africa were wage workers.¹⁰ In British colonies, too, various forms of coercion enabled the recruitment of African workers for colonial projects.¹¹

In 1930 the ILO’s Force Labor Convention imposed restrictions on the use of forced labor.¹² It was adopted by the ILO on 28 June 1930 and ratified by Britain on 3 June 1931 and France on 24 June 1937. Under pressure from representatives of colonized societies, France abolished forced labor in the colonies also by national legislation in 1946.¹³ Following the end of colonial rule, Convention 29 was ratified by independent West African countries in the 1960s. The Republic of Niger, where the case study discussed in this article is taken from, ratified it on 27 February 1961.¹⁴ While the ILO’s Forced Labor Convention reduced the coercive potential of labor recruitment, certain clauses in it allowed colonial empires to continue extracting labor coercively. Colonial developmentalism enabled the continued recruitment of unpaid native workers in the name of “community development” and “human investment” into local development.¹⁵ Rather than going from the abolition of forced labor to the introduction of free wage labor, colonial development schemes outsourced the coercive elements of labor recruitment to African chiefs and continued to mobilize African workers for free, or at a cost below the market value of labor. The use of developmentalist rhetoric to justify coercing people into volunteering their work lingered on after decolonization and was a well-documented strategy of the first African independent governments.¹⁶

The first decade that followed decolonization was characterized by a political discourse focused on nationalist and anticolonial modernization.¹⁷ It soon appeared that Africa’s economic development would not go through the same “stages” that some development economists saw as characteristic of Euro-America’s history.¹⁸ By the 1970s Africa started being seen as deviating from global trends toward proletarianization as defined in classical Marxist theory.¹⁹ Free wage labor had not become the norm in most African societies. Instead, small-scale self-sufficient peasants and herders, and workers involved in mutualist labor relations, remained prevalent.²⁰ Although throughout the twentieth century the proportion of wage workers grew in West Africa, when compared to other world regions it remained small. It decreased further following structural adjustment and the collapse of the formal sector in favor of an expansion of the so-called informal economy.²¹

Concurrently, the legacies of slavery continued to influence employment relations and hampered the formation of working-class consciousness. Abolitionism had not established itself as a dominant ideology in all regions of West Africa. In some West African societies, pro-slavery ideologies continued to exist and the biological descendants of slaves continued to be subordinate to slaveowners.²² By the early 1980s all West African countries had abolished slavery legally. By the first decade of the twenty-first century some West African countries had also criminalized it. But the official state law was slow to penetrate all areas and groups. It was particularly slow in the Sahara–Sahel, which contains scarcely populated regions minimally controlled and policed by underfinanced states. Legal pluralism and hybrid legal systems are the norm in West Africa. Contemporary state law evolved from legislations introduced under colonialism that imposed European legal concepts and institutions. Unsurprisingly, pre-existing norms and legal frameworks continued to influence the regulation of justice. Moreover, in some West African countries, the majority of the population is Muslim, but state law is non-religious.²³ Following independence most West African Muslims saw abolitionism as legitimate and supported the laws passed by their states to abolish slavery.²⁴ But some groups followed particular interpretations of Islamic law that were incompatible with both state law and Islamic law as interpreted by official Islamic authorities. They included terrorist groups that raided, kidnapped, and enslaved people, especially girls and women.²⁵ These have been publicly condemned by both Muslim and inter-faith religious leaders.²⁶

Since the 1970s slavery has been effectively undermined thanks primarily to the work of national anti-slavery activists.²⁷ And yet, ideologies that see slavery as legitimate in certain circumstances influence not only so-called terrorists, but also other groups. Descendants of slaveowners anachronistically classify persons of slave descent as “slaves” and occasionally force the latter to work for them or behave as their enslaved forebears.²⁸ These phenomena are impossible to quantify with the data available at present. But in areas where slavery continues to exist some employers never ceased to have a choice between employing enslaved or free workers. In the 2010s the Kayes region of Mali has been in the news for the violence and abuses perpetrated by descendants of slave-owning classes against persons seen as slave descendants, who resisted the former’s claims on their property and work.²⁹ In Islamic contexts female slavery is peculiarly resilient in the form of concubinage.³⁰ This, sometimes referred to as the “fifth wife” or “wahaya” phenomenon, is particularly well-documented for the Tahoua region of Niger, where the NGO Timidria has been supporting the female victims of a trade in girls whose mothers are considered “slaves.”³¹ Such girls are treated as the property of the mother’s owner by descendants of slaveowners and sold to wealthy men wishing to acquire sexually accessible women (concubines) beyond the four statutorily free wives legally permitted in Islam. Concubines are expected to provide domestic labor in the households they join. Testimonies by victims emphasize harsh working conditions and brutal treatment.³²

Tahoua: Slow Death of Slavery, Slow Birth of Wage Labor

Slave and forced labor co-existed in Ader since the beginning of the twentieth century. Farming was possible, but did not generate important marketable surpluses. Hausaphone settled villagers farmed valleys and sold any surpluses; Tuareg pastoralists and their dependents owned and sold camels, cattle, and livestock. But, unlike further south in the Sokoto Sultanate, there were no major plantations farmed by slaves.³³ Why hold slaves at all if the profits promised by the fruits of slave labor were marginal? Stephen Baier, Paul Lovejoy, and myself have provided interpretations of the peculiar functions of slavery at the edge of the Sahara: self-sustaining villages of slave status provided grain, herding, and hospitality required by semi-nomadic slaveowners in the course of their movements.³⁴ Slaves provided the labor necessary to feeding local populations and, if possible, producing a surplus in years of abundant or normal rainfall. They themselves were exchanged for cereals during droughts. The exchange of people, who needed to be fed, for rare and costly grains at times of drought, was an important safety net in a place constantly exposed to the threat of famine. Even outside famines, relatively small numbers of persons enslaved in Tahoua were sold across the sub-region. Tahoua was not a “slave reservoir,” like the Mandara mountains in what today is Northern Cameroon or the Guera region of Chad.³⁵ But particularly within Tuareg segmentary lineage systems, the sale and exchange of slaves across allied and collaborating units facilitated trans-Saharan connectivity.³⁶ In the 1920s and 1930s, relatively small numbers of enslaved women and their children were sold across commercially allied groups in what today are Mali and Niger.³⁷ Enslaved persons also worked in axes of long-distance trade controlled by Ader-based traders, such as the Agalawa studied by Abdulkarim Umar Dan Asabe.³⁸

Entry

At the beginning of our period, when the control of the French administration was not yet firmly established, entry into slavery took multiple forms. van der Linden’s classification distinguishes between two modalities of entry: physical compulsion and constrained choice. The former can be mediated through money, or not. Among examples of entry into slave labor mediated through money (or through other currencies and goods), the sale of enslaved persons is well attested in Ader. Slaves were exchanged for goods such as cereals or livestock. Seima and her daughter were sold “for some oxen” in 1913; Seidi, a ten-year-old girl, was sold for three heads of cattle and two donkeys in 1917; Elgeme was sold with her two children for thirty sheep and one donkey in 1921; Rhali, the ten-year-old son of an enslaved mother, was sold for 12,000 measures of millet in 1920; two Fulani boys were purchased for seven camels and put to work as herders in the mid-1930s.³⁹ I have interviewed one elderly lady who remembered having been kidnapped as a child and then sold for six male camels and one female camel at the market of Bilma in the 1930s.⁴⁰ I have not recorded cases of slaves-for-hire or slavery incurred for the payment of a debt. Tahoua’s respondents stated specifically that the former did not happen. However, both forms of entry (slaves-for-hire and debt-slavery) are attested in regions located south in the Sultanate of Sokoto.⁴¹

Among the forms of entry not mediated through money, people entered into slavery via abduction and birth to an enslaved mother, the most frequent modalities of entry into slavery in the first quarter of the twentieth century. Punctual kidnaps primarily targeted children, who were easily abducted if left unguarded by adults:

Once two Iwellemmeden [Tuareg] came to Charingué riding their camels. One of them grabbed a little child. But the child’s mother saw this, so she ran and grabbed the hand of the child. The man who held the child hit her on the head with his sword [takoba] but she wouldn’t let go. So the other […] told him to leave the child. He did… the child grew up here and died a few years ago.⁴²

Raids and kidnaps were frequent:

if [the Tuareg nomadic warrior elites] raided cattle, they took it away, then they stopped somewhere and every [participant in the expedition] could take some animals; if they raided captives, they did the same; if they raided a free man, he would ransom himself [by giving] money, cattle, or slaves.⁴³

Noteworthy in this quote is the distinction between persons of slave (“captive”) and free status. Communities considered to be of slave status, reproducing themselves biologically as slaves and living in hamlets owned collectively by particular individuals and families formed a social stratum at the bottom of Ader’s society. Tuareg warrior elites could raid, kill, pillage, or abduct people from their own slave groups or, more frequently, from slave groups tied to other Tuareg elites. While also free persons could be enslaved (and possibly ransomed), groups designated collectively as “slave” were frequent targets of re-enslavement. If slaveowners wanted to punish their enslaved dependents they could sell them; but they stood more to gain by acting as protectors toward their own slaves and kidnapping those enslaved to enemy groups.

Weaker, poorer, and more vulnerable than free-status groups, “slave” groups commonly sought the protection of those acting as their masters. Self-enslavement is attested, especially for enslaved persons known as bayun yunwa, slaves of famine or slaves of hunger, who were acquired in two ways: “If you had slaves but had nothing left to eat, you could sell one. Or some people who had nothing to eat could go to a rich person and offer themselves as slaves in exchange for food. Then these people would have to pay a ransom if they wanted to marry and be free.”⁴⁴ Persons threatened by famine, poverty, or facing wars, placed themselves under the protection of groups they thought of as powerful and thereby accepted a state of dependence equated with slavery. Self-enslavement is occasionally portrayed as a completely free choice. Even today there are persons who argue that they believe that it is God’s will that they serve a master whom they portray as quintessentially superior to them.

As official anti-slavery measures became progressively more effective, slave labor became a less common form of coerced labor throughout the twentieth century. But it did not die out entirely. From the perspective of persons who hold pro-slavery ideologies, the children of slave mothers who have not been manumitted by former masters can be legitimately enslaved.⁴⁵ In these circumstances, slavery is not only an illegal condition, but also a social status inherited across generations.⁴⁶ The descendants of slaves are classified as slaves. Classificatory slavery does not imply the presence of actual conditions of enslavement.⁴⁷ But the status of a slave is not inconsequential. Some classificatory slaves can choose how to live their lives, they are discriminated against and at risk of being coerced into conditions of actual enslavement. This possibility is concrete and explains why anti-slavery activists in the Sahel distinguish between active and passive slavery.⁴⁸

The notion of “passive slavery” may appear meaningless from the perspective of international law that identifies slavery with the presence of actual control tantamount to coercion.⁴⁹ But it is not meaningless in a society where classificatory slavery can be activated anytime. This happens, for example, when the children of statutory “slaves” are taken away by persons considering themselves their rightful owners and either sold or put to work as herders or, for girls, domestic workers. The most common form of entry into slavery in Tahoua today is the wahaya phenomenon.⁵⁰ The daughters of slave mothers are sold to men who seek a slave concubine. This form of entry into slave labor has continued to exist throughout the twentieth century. It is not until the famous case of Hadijatou Mani, who in 2008 won the case Hadijatou Mani vs Republique du Niger, that legal trials started taking place featuring victims of slavery denouncing enslavers with the support of anti-slavery NGOs.⁵¹

Let us turn to entry into forced labor. Under the six forms of entry by physical compulsion (sale, slaves-for-hire, payment of debt, birth, abduction, labor tax) only labor tax played a major role in the entry of Ader’s workers in colonial forced labor. Of the three sub-types of “constrained choice” (self-sale into slavery, money taxes, “free” wage labor), money taxes played a role. No one, to my knowledge, enrolled in colonial forced labor as a free choice. This is different from voluntary enslavement, because low-ranking persons expected greater returns from the protection of a loyally served master than from a colonial officer satisfied with the performance of native (forced) workers. Forced labor recruitment involved compelling workers to join worksites under threat of punishment. The colonial state organized public works. Needs in labor were identified and District Officers provided quotas to “traditional chiefs” (chefferie traditionelle), who recruited laborers. Often those selected were slaves and slave descendants.⁵²

In the 1920s–1940s the Indigénat served as the disciplinary mechanism for enforcing the administration’s orders.⁵³ “Prestations” referred to a specified number of days of unpaid labor that able-bodied men had to contribute to the administration of the Cercle. In the 1920s the French administration distinguished between prestations and the head tax. Prestatory labor was not expected to exceed 12 days per man, per year. When worksites were located further than 30 km away from the workers’ village, workers were fed or reimbursed the cost of their meals. The number of workers recruited varied depending on the cercle’s programme of works. Should there be no need to carry out public works, subjects would not be expected to contribute their labor in the form of prestations. Similar exemptions could not occur in the case of taxes, which were an obligation for all French subjects.

When France ratified Co29 in 1937, it had to modify the regime of prestations in conformity with the new legislation on forced labor. In the reasoning of colonial bureaucrats, a “radical solution” would have led to the abolition of prestations and the substitution of prestatory labor with voluntary workers paid with funds made available by “a new tax added on to the [head] tax.” But in the Colony of Niger, which was one of France’s poorest colonies, this would have resulted in excessive fiscal pressure. To avoid the impression that this was forced labor by another name, prestations continued as before, but now had to be regarded solely as a tax. Turning the earlier rhetoric on its head, official communications now argued that “The prestation must now be seen as a direct canton tax (impot direct cantonal), while we wait for the political and economic evolution of these regions to allow us to turn it into a direct communal tax. In principle, it is payable in one of two options, in cash or in labor, during the three months of tax collection.”⁵⁴ Seven years later, these points were still being reiterated in circulars:

The préstataires are men paying their taxes. They have no right to a salary. The only cost they should give rise to should be related to the provision of their meals when they work far from their village. . . . The levying of prestations is of a fiscal order . . . . The notion of forced labour imposed arbitrarily, which wrongly informs the conception that badly informed colonial [subjects] have of prestations, must be replaced by the notion of a just charge, proportional to the means of each and every person, and against which there can be no preconceptions.⁵⁵

In 1938, in the Cercle of Tahoua 14 percent of taxable men cleared their fiscal charges in cash, while the rest opted for prestations in labor, which amounted to between 2 and 10 workdays per man.⁵⁶ This central data suggests that coerced labor was not a heavy burden. However, correspondence from the cercles suggests otherwise. Locally based District Officers complained that prestations were onerous and that paying one’s taxes in kind (prestatory labor) or cash (the rachat option) was in competition with meeting subsistence needs. It is likely that a disproportionate amount of prestations were actually done by a small section of the taxable population, former slaves and poorer commoners, who were also the most vulnerable to famine, because their farms were smaller and on less productive land than those of people of free descent.

When in 1946 France abolished forced labor, it mobilized developmentalist rationales to justify paying indigenous workers half the market rate for their labor. Entry into colonial development projects followed two avenues: men who had limited opportunities to succeed as seasonal labor migrants or in other, more desirable, occupations, volunteered to join these initiatives. Others were forced by local chiefs and elites, themselves under pressure by colonial officers. Following independence in 1960 the regime of President Seyni Kountché, who had seized power through a coup, championed nationalist development projects and participants had no choice but to volunteer in the projects of the “Development Society.” Some men continued to migrate and farm for themselves, or work in the main axes of trade. But the national development worksites recruited workers through a mix of ideological and physical pressure.

The 1970s and 1980s were years of famine in the Sahel, and international development projects intervened in the Tahoua region. The Keita Project was a case in point. This Italy-funded, FAO-executed project started working at the time of Kountché’s regime and initially struggled to mobilize labor on its anti-desertification worksites. Men did not come forward, as a large proportion of the adult male population had migrated away from their famine-stricken villages. It was women who became the Keita Project’s workforce, remunerated in one ration of food per day of work.⁵⁷ By van der Linden’s criteria, these women counted as coerced labor. They freely chose to enroll into project work. But their choice was constrained by a lack of alternatives. Poor women of slave descent averted hunger and fed their children by working on the project.

Extraction

van der Linden conceptualizes extraction in terms of the incentives that employers provide to motivate coerced workers to work well and hard. He divides incentives into three sub-categories (compensation, coercion, and commitment), each of which is further internally subdivided into eleven sub-types.⁵⁸ I will only discuss forms of extraction relevant to Tahoua. The slave category was stratified internally into gradations of dependence. In theory if not always in practice, masters were responsible for their slaves’ basic needs. The most marginal slaves could be sold. These were recently captured slaves, who had not been integrated into the society of the masters. The two main slave categories were domestic slaves and allotment slaves. These two groups had different functions.⁵⁹ Domestic slaves, in Hausa bayun murfu (“slaves of the hearth”), took care of the everyday domestic needs of masters and followed their owners along their incessant travels, building their tents, looking after their cattle, taking care of their goods and families in their absence, and attending to the needs of their guests.⁶⁰ Allotment slaves lived in settled or semi-nomadic camps where their lives were relatively autonomous from those of their owners. While domestic slaves were fed and clothed by the masters, allotment slaves had to provide their own food, clothing, and shelter. The masters could take them away or appropriate their resources at any time. Domestic slaves could not move independently, for their residence and activities were tied to those of their owners. Allotment slaves, on the other hand, led a quasi-autonomous existence. Yet their mobility in space was restricted to the camp and its vicinities.

Also, various types of royal slaves existed, but domestic and allotment slaves were both the most common and most numerous categories of slaves, together with slave concubines, who were only female. Domestic slaves never received “wages” in the English sense of this term, but masters were obliged to support their domestic slaves’ basic needs in terms of food, clothing, and healthcare. This was the case even though domestic slaves were poorly fed and it was thanks to the collective work of slaves that masters were able to acquire food and wealth. Ideologies of slavery encouraged slaves to derive pride and a sense of self-worth from their loyalty toward masters and to serve the latter willingly. It is impossible to overestimate the pressure of these ideologies and, vice-versa, the fear that failing to serve a master might result in the latter’s curse and in God’s punishment.⁶¹

Extraction of slave labor also took the form of physical violence and torture. Beatings, castration, harassment, rape, taking away the enslaved’s most prized possessions and hurting their children, forcing the enslaved to carry out undignified tasks, and humiliating them in public: there is no category of violence that is not attested in regional research on slavery. Throughout the twentieth century all these forms of extraction continued, but violence became less common, both because slavery shrank as a form of employment and because violence was the behavior most likely to lead to official complaints. With the passing of anti-slavery legislation and with the criminalization of slavery, slave-owning could not be defended anymore. But especially in poorer hinterland regions, pro-slavery ideologies lingering on.

Turning to forced labor, minimal wages were provided to those forced to work far from their villages. These wages, while minimal, were valued by the formerly enslaved as a source of income that allowed them to start autonomous economic activities—from funding one’s own travels as labor migrants to purchasing goods that could be resold at some profit. The memory of forced labor is still alive in many villages and adds important details to the information available in colonial archives. Men from one village said they started migrating at the times of forced labor.⁶² The French made them cut down trees in Ader and carry them on foot to Tahoua, a two-day trip. They remained in Tahoua overnight and returned on the following morning. Sometimes nothing else was required from them for a long period until the village chief announced that he had to recruit more workers. Tasks varied. After carrying trees, they had to collect chalk from quarries. Those recruited spent a month working in the mines next to Bermo. When they were released, they were given a little money.⁶³ Some of them used it to finance their migration to Jos in Nigeria. This is how some elders remembered the organization of colonial worksites:

The French introduced taxation (‘ampho’, from the French ‘impot’)…. also forced labour (aikin dole), which consisted in taking people to worksites for 10–20 days. Workers were fed, but not paid [another speaker added: they were not fed either, they had to bring their own food!] They worked toward Tahoua, Gaya, even outside the country. The Lissawan [Canton chiefs] collaborated with the French, arranging the forced labour worksites. Each sector in the Canton had a supervisor, who worked for the Lissawan. . . . Alkassoum supervised the area of Tinkirana/Loudou. Attauri supervised the area of Waddey/Tabofat. Abuzeidi supervised the area of Tegueleguel. He had a helper in the village, who was designated by the village chief, but he came alone from Keita, he had no other assistants. […] Lissawan sector chiefs supervised forced labourers. They also collected taxes (bodu). On top of forced labour, people worked on the chief’s fields (gandun sarki) for one day at the heaviest stage of the farming season. For this they received a meal of porridge (fura). Abuzeidi could not speak Tamasheq, but the village chief (of Tegueleguel) spoke Hausa. Communication was not a problem. The Lissawan supervisor would just sit and oversee the works. He would select local people to check each worksite. There were no Frenchmen present, but soldiers visited to make sure that work proceeded smoothly. Workers grouped spontaneously with others like them: the Hausa with the Hausa, the Tamasheq with the Tamasheq. People from different villages tended to form separate work parties, if they could.⁶⁴

The Code de l’Indigénat was the legal instrument devised specifically to endow the colonial administration with arbitrary power to punish, and even execute, natives who did not perform according to their expectations. A letter from the Commandant of the Cercle of Tahoua written in 1940 exposes the administration’s attitudes toward ex-slaves (Bellah):

The Cercle’s populations are still far from a degree of evolution in which this regime [the Code de l’Indigénat in its 1887 form] would be considered excessive, and immediate sanctions are always more effective against primitives than penalties that sometimes may not occur until after a fortnight has passed [from the perpetration of an infraction of the code] . . . infractions are extremely frequent particularly amongst the ex-slaves (Bellah) who have become very independent now that they are not enslaved anymore and, having escaped the stick of the master, would pretend to evade all authority.⁶⁵

With the spread of developmentalist discourses, which gradually replaced previous forms of labor extraction, wages, and commitment became more prominent incentives, replacing coercion and violence. Pride in the results of work carried out in the name of local and national development is central to developmentalist ideologies, as is shame for failing to perform successfully. However, failure to meet targets and achieve results could lead to punishment, such as denial of support by rural extension officers or, occasionally, land expropriation.

Exit

van der Linden lists seven variants of exit from chattel slavery: “owners can compel slaves to leave, or they can be forced to free them by another entity (an abolitionist state, for example), but they can also free them unconditionally; slaves can run away or mutiny; they can purchase their own freedom; they can remain as slaves; or they may die.”⁶⁶ All these circumstances are attested to Tahoua. In Tuareg societies, forced exit occasionally took the form of disappearance of slaveowners, especially for allotment slaves. Certain masters’ families died out, such as those who resisted French occupation and were exterminated; or became so impoverished that they severed social ties and lived as poor nomads isolated from dependents. Others lost their livestock in famines and had to rely on the help of former slaves. The latter configuration eludes easy classification in clear-cut typologies. Some former masters used ideological threats to convince former slaves to continue serving them. But I have also collected testimonies in which ex-slaves explained that they had continued helping the descendants of former masters out of genuine pity.

Up until the 1920s exit from slavery happened mainly through manumission, escape, or self-ransom. Then, from the 1920s onward, colonial abolition started being systematically enforced. Slaves who wished to distance themselves from masters either left (this option was prevalent among young men of slave descent) or denounced the bad treatments of masters to colonial authorities, an option prevalent among women who were more often traded or faced the pain of having their children taken away from them. Since the famous case of Hadijatou Mani, more concubines came to the fore to denounce their own sale and unwanted servitude through recourse to NGOs and national justice. Exit dynamics are highly gendered; it is rare for slaveowners to attempt to stop male slaves from breaking ties of dependence.

The notion of exit from slavery suggests a total break. But in Tuareg society individual slaves, or entire slave villages, could not be completely liberated from their servile status. If they were manumitted, their prior enslavement continued to influence their status as they were reclassified as liberated slaves. Freed slaves fell into two categories, the ighawellan, who had enjoyed free status for a long time, and the more recently freed iderfan. They could not become assimilated into freeborn people (ilellan). Slaves or ex-slaves were seen as ethnically different from “real” Tuareg and denied full human status by members of the elite. Racialized criteria differentiated them from freemen also after manumission.⁶⁷ This is one of the reasons why a large number of people of slave descent changed ethnicity and Hausaised as an emancipation strategy. This process lasted two or three generations and involved moving to a Hausa village and adopting the Hausa language and lifestyle.⁶⁸ To be sure, slavery existed in the Hausa-speaking societies of Ader, but social mobility was easier here than in Tuareg contexts.

Exit from colonial forced labor took primarily the form of resistance and escape. Ader’s workers resisted so vehemently to forced enrollment in colonial worksites that the colonial administration hesitated to impose recruitment in this region. This was a consequence of the limited potential of this region as a source of colonial revenue. Local workers became seasonal labor migrants both to pay, and to avoid paying, colonial taxes. Their migration toward British Nigeria was a major concern to French administrators. Ader’s potential employers competed against the revenues that could be derived by migrating, which were almost invariably higher than what employers would be willing to offer workers in Ader.

What was produced in Ader faced barriers to commercialization in the form of high transport costs. Only few private producers/traders in local cash crops employed local workers who could not or would not migrate. The colonial and independent state could try to coerce workers to work by impeding their departures. But enforcing a systematic and effective surveillance of laborers was not worth the cost of coercion. Due to ideologies that stopped them from migrating, women were more likely to accept the low rates of remuneration that some employers would offer—as in the case of the Keita Project. Elderly women of slave descent are those most likely to continue serving former masters—their weakness and limited alternative livelihood options make subservience more acceptable.

From the 1920s onward, ex-slaves who owned scarcely fertile lands on northern Ader’s rocky slopes turned into seasonal migrants, not into self-sufficient peasants. They diversified their livelihoods: women and children farmed the family’s dry fields, and men worked for wages in West Africa’s cities. Some of the poorest groups migrate locally, often with the whole family, following a particular form of migration known as “cin rani,” which involves leaving one’s household after the harvest until the following rainy season when people come back to farm and rely on the previous harvest’s stocks at a time when the price of cereals is highest. Those practising this type of migration, even today, are often persons of slave descent who migrate locally to work for descendants of former slaveowners in the dry season. Women carry out domestic work for their employers during the day, and they take two meals (with their young children, if they had any) at their employers’ homes. They are paid the equivalent of 5–10 GBP per month. Husbands might work for the same employers; or work abroad as long-distance migrants; or work in a larger village nearby where jobs are available in local trade, irrigated agriculture, or the service sector.

Conclusion

Coercion is costly. Coercers must invest resources and energy into forcibly recruiting people; supervising workers and providing incentives to make them work hard; and preventing them from exiting their exploitative circumstances. Coercion is also costly on moral grounds. The exploitation of particular individuals or groups must be justified morally and politically. This requires convincing people that coercion is somehow legitimate according to society’s moral compass. Throughout the twentieth century, indigenous slavery and colonial forced labor were progressively contested, and eventually outlawed, because people changed their views about which forms of coercion were legitimate and which ones weren’t. Different groups changed how they thought about various forms of coercion at different moments.

The colonizers considered themselves abolitionists and presented the eradication of traditional slavery as a moral justification for the imposition of their rule. They introduced new forms of forced labor. Until the mid-1940s, French administrators spoke of the compulsory work they were imposing on African societies as a way to turn Africans into “free” workers. But eventually, thanks mainly to African protests and resistance, French policymakers and administrators came to see the forced labor they had imposed on African colonial subjects as unjust. Concurrently, in the early twentieth century, many West Africans saw slavery as a legitimate institution and resented colonial forced labor and colonialism. They progressively changed their views about slavery. The first to resist slavery were the enslaved. They also mobilized against colonial rule and forced labor. Ideas changed, but the Sahel’s environment never ceased to threaten potential employers and employees alike with high risks of business failure and low returns to commercial agriculture.

Comparing slavery and forced labor in this context shows that slavery has been more resilient not because it maximized anyone’s profits, but because it had potential to reduce risk. This was true for both slave-owning and enslaved groups, but the nature of their relationship was far from mutualistic: slaveowners dominated those enslaved and reaped potentially greater benefits from slavery than the enslaved, who would not resist enslavement when it protected them from hunger. But when hunger and extreme poverty were not unavoidable, as I showed elsewhere, they sought freedom.⁶⁹ Poverty is unavoidable for people who, for whatever reason, cannot travel and who cannot access adequate food by working for themselves or for others. By van der Linden’s definition, working for others is always coercive except when the arrangements can be characterized as mutualistic. In general, Ader’s circumstances hampered the ability of potential employers to offer wages. Why, then, employ dependent labor at all, after abolition?

Under colonial regimes, the French administration had to prove that it could govern the population of these Sahelian regions, and this involved mobilizing their labor. During the Sahel crisis, international development projects whose vocation it was to stop the desert’s advance had to enroll the local workforce in the anti-desertification worksites. This made it necessary to mobilize local labor. But it mattered relatively little whether this labor achieved its ends lastingly and efficiently. Finally, Ader does contain limited fertile lands that produce onions and tomatoes. These are local cash crops and a small group of entrepreneurs offer wages to those willing to work on their farms. Many of the latter are local small-scale farmers, whose family members work on their own farms and who also borrow seeds and inputs from their employers on credit and pay them back in kind at the harvest. Poverty and debt account for the coercive conditions of statutorily “slave” and “free” labor at the desert’s edge.

van der Linden’s analytical framework is useful. It makes it possible to compare coercive labor regimes without using concepts like “slavery” that carry heavy moral connotations. In Euro-America, we have come to see slavery primarily as the worst form of labor commodification. van der Linden’s framework allows researchers to compare different forms of labor coercion with greater precision, while eluding the ideological straitjackets implicit in our own analytical language. But even when “exit, extraction, and exit” are applied, most of the work of interpretation still remains to be done. The three moments framework is no magic formula. It does not in itself answer questions like why has slavery proven more resilient in certain contexts than in others.

Some of van der Linden’s assumptions require more reflection. In particular, the idea that all labor for others should be seen as coerced labor is a simplification that would benefit from some conceptual nuancing. How can the intensity of coercion be theorized for comparative purposes? Being enslaved through abduction in a violent raid on one’s village is experienced differently from enrolling into an international development project for a daily meal as a wage but no other pressure by the employer. The threat of hunger and the bitter choices it forces one to make are experienced differently from the physical pain inflicted by an abductor or a violent master. How to account, analytically, for these differences? Noel Lenski’s suggestion to develop an indicator—he calls it “vectors of intensification”—to measure the intensity of exploitation in coercive relations is helpful.⁷⁰ Such a measurement would make it possible to weigh benefits to masters (use and exchange value of slaves as commodity, or of slave labor and its fruits) against disadvantages to enslaved persons (permanence of the slave condition, level of violent domination, degree of natal alienation and dishonor). This matters not only because it might increase the clarity and precision of our understanding of the past, but also because it can inform decisions on how to give more and better choices to potential victims of coercion in the future.

Notes

• 1. Research for this article received funding from the European Research Council (ERCAG 885418, RISE 734596) and the Arts and Humanities Research Council of the United Kingdom (AH/R005427/1, AH/V01210X/1) and the author wishes to thank these institutions for their generous support.

• 2. Eric Williams, Capitalism and Slavery (Chapel Hill: University of North Carolina Press, 1994, first edition 1944); Seymour Drescher, Econocide: British Slavery in the Era of Abolition (Chapel Hill: University of North Carolina Press, 2010, first edition 1977); Howard Temperley, “Capitalism, Slavery, and Ideology,” Past and Present 75 (May 1977): 94–118; David Eltis and Stanley Engerman, “The Importance of Slavery and the Slave Trade to Industrializing Britain,” Journal of Economic History 60, no. 1 (2000): 123–44.

• 3. Gareth Austin, “Capitalists and Labour in Africa,” in General Labour History of Africa: Workers, Employers and Governments, 20th–21st Centuries, ed. Stefano Bellucci and Andreas Eckert (Cambridge: Cambridge University Press, 2021), 425–56, here 436.

• 4. Testimony of Dela Mahamadou, interviewed by Apsatou Bagaya and Benedetta Rossi in Niamey, 13 November 2021.

• 5. Note on the terminology of slavery: the terms “slave” and “master” are used to indicate socially recognized statuses that feature in the sources referred to in this chapter. Despite colonial and national abolition laws, some groups continue to view “slave” and “master” statuses (and related living conditions) as legitimate. These terms carry normative connotations and reveal the persistence of non-abolitionist ideologies. This chapter interrogates and analyzes the causes of this persistence. Today most anti-slavery activists in the West African Sahel reject the use of the term “enslaved person” or “enslaver” as substitutes for the terminology that is actually used by people in these contexts. Some of these activists call for the term “slave” to be used as a way to denounce the continued existence of actual slavery in their societies, a terrible reality that they do not wish to euphemize, but choose to denounce and fight against. As this chapter seeks to show, the circumstances are complex and all terminological choices are problematic. The use of the term “slave” in this chapter is meant to retain the sources’ terminology that reveals the endurance of a pro-slavery discursive regime. It is never meant to imply a pejorative connotation: it refers to a status/condition which should have disappeared, but which continues to exist in practice and in local parlance. This attests to the ongoing oppression of the groups and individuals labeled “slaves” (some of whom may face actual conditions of enslavement) by the groups and individuals labeled “masters.” It attests, too, to the presence of local activists who struggle against slavery and its legacies.

• 6. Gareth Austin, “Cash Crops and Freedom: Export Agriculture and the Decline of Slavery in Colonial West Africa,” International Review of Social History 54, no. 1 (2009): 1–37.

• 7. Frederick Cooper, “From Enslavement to Precarity? The Labour Question in African History,” in The Political Economy of Everyday Life in Africa: Beyond the Margins, ed. Wale Adebanwi (Oxford: James Currey, 2017), 135–56.

• 8. Marcel van der Linden, “Dissecting Coerced Labor,” in On Coerced Labor: Work and Compulsion after Chattel Slavery, ed. Marcel van der Linden and Magaly Rodriguez Garcia (Leiden: Brill, 2016), 293–322, here 294. See also Christian De Vito, Juliane Schiel, and Matthias van Rossum “From Bondage to Precariousness. New Perspectives on Labor and Social History,” Journal of Social History 54, no. 2 (2020): 1–19.

• 9. Martin Klein, Slavery and Colonial Rule in French West Africa (Cambridge: Cambridge University Press, 1998).

• 10. Frederick Cooper, Decolonization and African Society: The Labor Question in French and British Africa (Cambridge: Cambridge University Press, 1996), 31; Babacar Fall, Le travail forcé en Afrique Occidentale française (1900–1945) (Paris: Karthala, 1993), 17–19; Romain Tiquet, Travail forcé et mobilisation de la main-d’œuvre au Sénégal, Années 1920–1960 (Rennes: Presses Universitaires de Rennes, 2019).

• 11. Cooper, Decolonization and African Society, 58–73.

• 12. The full text of the ILO’s Forced Labor Convention (Co29) can be found at the following link (checked and working on July 8, 2017), http://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:12100:0::NO::P12100_ILO_CODE:C029.

• 13. Fall, Le travail forcé.

• 14. For a discussion of the circumstances in which the Convention was developed, see Suzanne Miers, Slavery in the Twentieth Century: The Evolution of a Global Problem (Walnut Creek, CA: Altamira, 2003), 134–51.

• 15. Benedetta Rossi, “What ‘Development’ Does to Work,” International Labor and Working Class History 92 (Fall 2017): 7–23.

• 16. Babacar Fall and Richard Roberts, “Forced Labor,” in General Labour History of Africa: Workers, Employers and Governments, 20th–21st Centuries, ed. Stefano Bellucci and Andreas Eckert (Cambridge: Cambridge University Press, 2021), 77–115, here 112.

• 17. Corrie Decker and Elisabeth McMahon, The Idea of Development in Africa: A History (Cambridge: Cambridge University Press, 2021), 143–63.

• 18. Frederick Cooper, “African Labor History,” in Global Labour History. A State of the Art, ed. Jan Lucassen (Bern: Peter Lang, 2006), 91–116.

• 19. John Sender and Sheila Smith, The Development of Capitalism in Africa (London: Taylor & Francis, 1986).

• 20. Karin Hofmeester, Jan Lucassen, and Filipa Ribeiro da Silva, “No Global Labor History without Africa: Reciprocal Comparison and Beyond,” History in Africa 41 (2014): 249–76, here 268. Marcel van der Linden defines mutualism as “all voluntary arrangements, in which people make contributions to a collective fund, which is given, in whole or in part, to one or more of the contributors according to specific rules of allocation.” Marcel van der Linden, Workers of the World: Essays toward a Global Labor History (Leiden: Brill, 2000), 81.

• 21. Structural Adjustment Participatory Review International Network (SAPRIN). Structural Adjustment—The SAPRI Report: The Policy Roots of Economic Crisis, Poverty and Inequality (London: ZED Books, 2004); Béatrice Hibou, ed., La privatisation des états (Paris: Karthala, 1999).

• 22. Ibrahima Thioub, “Regard critique sur les lectures africaines de l’esclavage et de la traite atlantique,” in Les historiens africains et la mondialisation, ed. Issiaka Mandé and Blandine Stefanson (Paris: Karthala, 2005), 271–92.

• 23. This is the case of Senegal, Niger, Mali, Chad, and Guinée, for example. By contrast, in Mauritania and some states in Northern Nigeria shariah law is national law.

• 24. William Gervaise Clarence Smith, Islam and the Abolition of Slavery (London: Hurst & Company, 2006); Khaled Abou el-Fadl, The Great Theft: Wrestling Islam from the Extremists (Harper San Francisco, 2005); Bernard K. Freamon, “Slavery, Freedom, and the Doctrine of Consensus in Islamic Jurisprudence,” Harvard Human Rights Journal 11, no. 1 (1998), accessible at https://ssrn.com/abstract=1564321.

• 25. Lee Ferran, Dana Hughes, James Gordon Meek, and Colleen Curry, “Boko Haram: Kidnappers, Slave-Owners, Terrorists, Killers,” ABC News, 5 May 2014, accessible at https://abcnews.go.com/Blotter/boko-haram-kidnappers-slave-owners-terrorists-killers/story?id=23598347.

• 26. In 2014, Iraqi-Iranian Shia political-religious leader Ayatollah Mohammad Taqi al-Modarresi met with the Leader of Catholic Christianity Pope Francis and other religious authorities to draft an inter-faith declaration to “eradicate modern slavery across the world by 2020 and for all time.“ The declaration was signed by several Shi'ite leaders including the Sunni Grand Imam of Al Azhar Mosque in Cairo (Egypt). See Global Freedom Network, A United Faith Against Modern Slavery: The Joint Declaration of Religious Leaders against Modern Slavery. Vatican City, 2 December 2014, https://cdn.walkfree.org/content/uploads/2020/10/06154604/GFN-2-Dec-Vatican-Book-digital.pdf. Muslim scholars also released a separate open letter to condemn the declarations and activities of ISIS and Boko Haram in 2014, see: Religion News Service, “Muslim Scholars Release Open Letter To Islamic State Meticulously Blasting Its Ideology,” The Huff Post, 24 September 2014, https://www.huffpost.com/entry/muslim-scholars-islamic-state_n_5878038.

• 27. Eric Hahonou and Lotte Pelckmans, “West African Antislavery Movements: Citizenship Struggles and the Legacies of Slavery,” Stichproben: Wiener Zeitschrift für kritische Afrikastudien 20 (2011): 141–62; Eric Hahonou, “Slavery and Politics: Stigma, Decentralisation, and Political Mobilisation in Niger and Benin,” in Reconfiguring Slavery: West African Trajectories, ed. Benedetta Rossi (Liverpool: Liverpool University Press, 2009), 152–81; Zekeria Ould Ahmed Salem, “Bare-Foot Activists: Transformations in the Haratine Movement in Mauritania,” in Empires of Speed: Time and the Acceleration of Politics and Society, ed. Robert Hassan, Stephen Ellis and Ineke van Kessel (Leiden: Brill, 2009).

• 28. Benedetta Rossi, ed., Reconfiguring Slavery. On the political and identitarian reconfigurations of relations rooted in slavery, see also Baz Lecocq, “The Bellah Question: Slave Emancipation, Race and Social Categories in Late Twentieth-Century Northern Mali,” Canadian Journal of African Studies 39, no. 3 (2005): 42–67.

• 29. Lotte Pelckmans, “Breaking the Silence? The Ongoing Legacies of Internal Slavery in Mali,” article published online in August 2000, accessible at https://www.slaveryforcedmigration.org/en/article-breaking-the-silence-the-ongoing-legacies-of-internal-african-slavery-in-mali-lotte-pelckmans-aout-2020-en-anglais/.

• 30. See, for example, Ann McDougall, “A Sense of Self: The Life of Fatma Barka,” Canadian Journal of African Studies 32, no. 2 (1998): 285–315.

• 31. Moussa Zangaou and Kadir Abdelkader Galy, Wahaya, Domestic and Sexual Slavery in Niger (Niamey: Antislavery International and Timidria, 2012), https://www.antislavery.org/wp-content/uploads/2018/10/Wahaya-report.pdf.

• 32. Testimony of Dela Mahamadou, interviewed by Apsatou Bagaya and Benedetta Rossi in Niamey, 13 November 2021.

• 33. Mohammed Bashir Salau, The West African Slave Plantation: A Case Study (London: Palgrave Macmillan, 2011); Mohammed Bashir Salau, Plantation Slavery in the Sokoto Caliphate: A Historical and Comparative Study (University of Rochester Press, 2018).

• 34. Stephen Baier and Paul Lovejoy, “The Desert Side Economy of the Central Sudan,” The International Journal of African Historical Studies 8, no. 4 (1975): 551–81; Stephen Baier and Paul Lovejoy, “The Tuareg of the Central Sudan: Gradations in Servility at the Desert’s Edge (Niger and Nigeria),” in Slavery in Africa: Historical and Anthropological Perspectives, ed. Suzanne Miers and Igor Kopytoff (Madison, WI: University of Wisconsin Press, 1977); Benedetta Rossi “Kinetocracy: The Government of Mobility at the Desert’s Edge,” in Mobility Makes States: Migration and Power in Africa, ed. Joel Quirk and Darshan Vigneswaran (Philadelphia: University of Pennsylvania Press, 2015), 223–56.

• 35. Valerio Colosio, “‘The Children of the People’: Integration and Descent in a Former Slave Reservoir in Chad,” Ph.D. Thesis, University of Sussex (UK), Department of Social Anthropology, 2018.

• 36. On trans-Saharan connectivity, see James McDougall and Judith Scheele, eds., Saharan Frontiers: Space and Mobility in Northwest Africa (Bloomington: Indiana University Press, 2012).

• 37. Benedetta Rossi, “Periodizing the End of Slavery: Colonial Law, the League of Nations and Slave Resistance in the Nigerien Sahel, 1920s–1930s,” Annales (Histoire, Sciences Sociales—English edition) 72, no. 4 (2017): 605–40, here 622.

• 38. Abdulkarim Umar Dan Asabe, Comparative Biographies of Selected Leaders of the Kano Commercial Establishment, MA Thesis, Bayero University (Kano, Nigeria), Department of History, 1987.

• 39. Data from Archives Nationales du Niger (ANN) 381; if the measure is the tiya, 12,000 measures of millet correspond roughly to 24 tons.

• 40. Her story is told in Rossi, “Periodizing the End of Slavery,” 635.

• 41. Murgu, in the Hausa-speaking world, is an institution through which a slave could pay money to his owner in lieu of the work owed to him. Some slaves were allowed to work for a salary but had to give all or part of that salary to their owners. On how this institution worked in the late nineteenth century and was transformed under colonial rule in Northern Nigeria, see Paul Lovejoy, “Murgu: The Wages of Slavery in the Sokoto Caliphate,” Slavery & Abolition 14 (1993): 168–85.

• 42. Interview with anonymized speaker, Charingue Tajae, 10 June 2005. Unless otherwise specified, all interviews were conducted in Hausa by the author with the assistance of Ahmadou Seydou.

• 43. Testimony of Alesmaghil Ag Mousa, in Francis Nicolas, “Textes ethnographiques de la Tamajeq des Iullemmeden de l’est, (quatrième partie),” Anthropos 51 (1956): 129–56, here 148–9.

• 44. Interview with Keitawa elders in Keita, 8 September 2005.

• 45. I discuss examples of these circumstances in “Beyond the Atlantic Paradigm: Slavery and Abolitionism in the Nigérien Sahel,” Journal of Global Slavery 5, no. 2 (2020): 238–69.

• 46. Valerie Couillard, Difficile Passage vers la liberté. Dix années de travail contre l’esclavage par ascendance en Afrique de l’Ouest (London: Anti-Slavery International, 2019).

• 47. Benedetta Rossi, “Introduction: Rethinking Slavery in West Africa,” in Reconfiguring Slavery, 5.

• 48. Galy Kadir Abdelkader, Slavery in Niger: Historical, Legal and Contemporary Perspectives (London and Niamey: Anti-Slavery International and Association Timidria, 2004), 9–10.

• 49. Jean Allain and Robin Hickey, “Property and the Definition of Slavery,” International and Comparative Law Quarterly 61 (2012): 915–38, here 923.

• 50. Moussa Zangaou and Kadir Abdelkader Galy, Wahaya.

• 51. Jean Allain, “Hadijatou Mani Koraou v. Republic of Niger: Judgement no. ECW/CCJ/JUD/06/08,” The American Journal of International Law 103, no. 2 (2009): 311–17; Helen Duffy, “Hadijatou Mani Koroua v Niger: Slavery Unveiled by the ECOWAS Court,” Human Rights Law Review 9, no. 1 (2009): 151–70.

• 52. Roman Tiquet, “Challenging Colonial Forced Labor? Resistance, Resilience, and Power in Senegal (1920s–1940s),” International Labor and Working Class History 93 (2018): 137; Ash, Catherine B. Ash “Forced Labor in Colonial West Africa,” History Compass 4, no. 3 (2006): 402–06.

• 53. Gregory Mann, “What was the indigénat? The ‘Empire of law’ in French West Africa,” Journal of African History 50, no. 3 (2009): 331–53; Alice Conklin, A Mission to Civilize: The Republican Idea of Empire in France and West Africa, 1895–1930 (Stanford: Stanford University Press, 1997).

• 54. Gouverneur Niger to Commandants de Cercles and Subdivisions, Circular, 12 March 1931, Archives Nationales du Senegal (hereafter ANS) K263 (26).

• 55. Gouverneur Niger to Commandants de Cercles, Circular No. 152, 30 July 1938.

• 56. Note for the Director of Political Affairs, 15 June 1939, ANS K263 (26).

• 57. Benedetta Rossi, “From Unfree Work to Working for Free: Labor, Aid, and Gender in the Nigerien Sahel, 1930–2000,” International Labor and Working Class History 92 (Fall 2017): 155–82.

• 58. van der Linden, “Dissecting Coerced Labor,” 308, Fig. 13.2.

• 59. This distinction, based on my enquiries in Ader, matches the observations of Bernus, “L’évolution récente des relations entre éleveurs et agriculteurs en Afrique tropicale: L’example du Sahel nigérien,” Cahiers ORSTOM 9, no. 2 (1974): 137–43, here 142; see also Edmond Bernus and Susanne Bernus, “L’évolution de la condition servile chez les Touaregs sahéliens,” in L’Esclavage en Afrique précoloniale, ed. Claude Meillassoux (Paris: Maspero, 1975), 27–47, 33–5.

• 60. Testimony of Rabi ag Elibdar in Francis Nicolas, “Textes ethnographiques de la Tamajeq des Iullemmeden de l’est, (troisième partie),” Anthropos 50 (1955): 635–58, here 655–6.

• 61. For a study of the power of masters’ curse, see Christine Hardung, “Curse and Blessing: On Post-slavery Modes of Perception and Agency in Benin” in Reconfiguring Slavery, 116–39.

• 62. Interview with Mousa Bukoci and other elders in Jiggina, 1 December 2008.

• 63. According to decree No. 184 of 31 March 1927, casual workers would receive a daily fee of 0.50 francs, which was the estimated cost of their evening meals. According to decrees No. 630 of 14 November 1927 and No. 34 of 17 January 1928, permanent workers recruited at the cercle level to carry out construction work for the colony would receive 100 francs after six months of work, and 200 francs after one year. Workers would receive one daily meal on the worksite, which decree No. 52 of 15 April 1928 increased from 125 to 200 grams of meat. Gouverneur Niger to Gouverneur Général, 1 May 1928, ANS K59 (19). It is, of course, not always clear if these legal provisions were actually implemented.

• 64. Interview with elders of Tegueleguel Teguef, 7 June 2005.

• 65. Commandant de Cercle of Tahoua to Governor of Niger, 22 May 1940.

• 66. van der Linden, “Dissecting Coerced Labor,” 314 and 311, Fig. 13.3.

• 67. Edmond Bernus and Susanne Bernus, “L’évolution de la condition servile,” 31–32.

• 68. I provided a detailed example of this process in “Being and Becoming Hausa in Ader (Niger),” in Being and Becoming Hausa: Interdisciplinary Perspectives, ed. Anne Haour and Benedetta Rossi, 113–40.

• 69. Benedetta Rossi, “Dependence, Unfreedom, and Slavery in Africa: Toward an Integrated Analysis,” Africa 86, no. 3 (2016): 571–90.

• 70. Noel Lenski, “Framing the Question: What is a Slave Society?,” in What is a Slave Society? The Practice of Slavery in Global Perspective, ed. Noel Lenski and Catherine Cameron (Cambridge: Cambridge University Press, 2018), 220–50, here 52–53.