In the scholarly argument over whether slavery was primarily a social, economic, or political phenomenon, Mier and Kopytoff take a benign view. They argue that slavery mainly served to assimilate slaves into the slave masters’ households as kinsmen. Hopkins disagrees. Building on Nieboer and Domar, he points to the use of coercion, and stresses that in the context of land abundance, labor scarcity, and simple agricultural technology, “the costs of acquiring and maintaining slaves were less than the costs of hiring labour.” Hopkins's approach is therefore centered on economic choice, and it assumes that slavery was primarily an economic institution: that the essence of the slave's role mainly resided in his labor, and that slaves were chattel. Meillassoux's contribution to the debate over the nature of slavery in Africa recognizes that slavery could be, but was not always necessarily, primarily a political institution, and that the basis of slavery was violence. It also stresses confl icts between masters and slaves, as well as an emphasis on the organization of production. For Klein and other scholars, both Hopkins's and Meillassoux's approaches are most appropriate for high-density slave systems such as the Sokoto caliphate. The evidence confirms that Klein and others are right. In particular, it shows that most slaves were enslaved by violence, that they were considered property, that they were not always treated mildly, and that they often resisted against their subordination. It also reveals the economic dimensions of slavery, by stressing that slaves were used for agricultural production and other economic tasks, that they were viewed as property, and that slaves who worked on their own account typically paid their masters significant amounts, which contributed to their inability to purchase their own freedom.

Although Hopkins's and Meillassoux's arguments can be extended to the Sokoto caliphate, one question remains unanswered or has not been adequately addressed by scholars: why in the Sokoto caliphate and other such high-density slave systems, unlike in many other parts of Africa, we do not find the assimilation of most plantation slaves into the masters’ households as kinsmen, or (in other words) how was the development of serfdom prevented?

With these words, Walter Scheidel began his interesting recent article on the expansion and nature of slavery in Classical Greece and Republican Rome. Read at its broadest, though dealing with very similar issues largely related to the development and nature of slavery in highdensity slave systems, as well as with the theme of comparative slavery, the present work, rather than focusing primarily on slavery in Classical Greece and Republican Rome, primarily focuses on plantation slavery in the largest Islamic state of nineteenth-century West Africa: the Sokoto caliphate.

The Sokoto caliphate (1804–1903) was situated in the central Sudan region of West Africa. It comprised two ecological zones: the Sudan savanna and the Guinea savanna. The former covered the northern parts of the caliphate, while the latter prevailed in the southern parts of the state. The soil types in the Guinea savanna support the production of diverse crops such as yam, cassava, guinea corn, cotton, tobacco, soy beans, cowpeas, and rice, while those in the Sudan savanna are more favorable for the production of cereals (millet, sorghum, maize, and rice), cotton, groundnuts, and legumes, as well as for livestock rearing.

The caliphate was established through a jihad led by a Fulani cleric known as Usman dan Fodio at the beginning of the nineteenth century. Hausaland was the principal theater of the jihad. Before the jihad, Hausaland consisted of several centralized Hausa governments or city states that were organized around markets, palaces, towns, and “closely settled zones.” These Hausa states had a long history of involvement in international trade, and by 1500 most of their rulers had embraced Islam. On the eve of the jihad, the Hausa states were ruled by nominal Muslims, and they were at war with each other.

The Sokoto caliphate covered a vast expanse of land that consisted of former Hausa states such as Gobir and Zamfara. Plantations could be found in most if not all of these states, as well as in other parts of the central Sudan, in both the Sudan savanna and Guinea savanna belts, even before the nineteenth-century Sokoto jihad. But the jihad and its aftermath encouraged the rapid expansion of the existing plantation sector, at least in the region that became known as the Sokoto caliphate; without the holy war, plantation expansion probably would not have been as notable.

From the earliest studies of the problem of slavery scholars have tried to explain how slavery, as an institution, developed. Nieboer's explanation linked land abundance to slavery, and he stressed that the use of slaves in societies having abundant land was a means of enhancing agricultural production or economic output in general. Following Nieboer, some scholars have identified demographic conditions of low population-to-land ratios in Africa, and noted that the availability of land that people could fl ee to in order to avoid working for others resulted in efforts to check labor mobility, and ultimately to the reliance on plantation slavery and other forms of coerced labor. Put differently, they, like Nieboer, noted that slavery was exclusively a response to economic incentives associated with land availability. Other scholars argue that land-labor pressures alone cannot explain how slavery and other forms of unfree labor developed. In so doing they stress that the development or cause of slavery in Africa was mainly tied to social or political factors.

Although we do not have data like Austin and other scholars had, to determine whether or not slave labor was cheaper than other forms of labor, clearly land abundance characterized pre-jihad Hausaland, and clearly plantation slavery was in the economic best interest of slaveholders, as chapter 6 of this book will demonstrate. Given these two facts, the present chapter accepts that plantation slavery in the Sokoto caliphate was partly a response to economic incentives associated with land availability. However, it argues that the origin of plantation slavery there was more closely associated with political policy, particularly with Bello's interest in economic and political consolidation, as well as with his interest “with the maintenance of an active front line for defense and annual campaigns.”

The Sokoto jihad was spearheaded by men of vision who were mainly concerned with the establishment of a just, ideal, and viable Islamic state modeled after that centered in Arabia in the seventh century. The consensus among these men was that Muslims are destined to rule over unbelievers, and that trade in enslaved Muslims was wrong. Among other factors, this resulted in the use of military action to support their vision. The nature and course of the resulting military engagements led to increased availability of land and slaves, which in turn fostered plantation development.

State Policy Goals, Innovations, and Plantation Development

Although the policy goals pursued by the Sokoto jihad leaders were in many respects consistent with those of the various Hausa regimes they displaced, they differed significantly, in part because they were primarily shaped by Islamic principles, and in part because the basis for these policies was the Muslim ummah, or broader Islamic community, rather than the limited regional interests which had underpinned the policies of the old regimes. It is not necessary to examine the pre-jihad and the post-jihad policy goals in detail. However, it is crucial to state that, unlike the pre-jihad leaders, their successors extensively documented their concerns. They wrote on diverse issues such as education, personal integrity, finance, and administration. The jihad leaders, however, were primarily concerned with the establishment of a just, viable, and ideal Islamic state modeled after the historic state centered in Arabia in the seventh century. In conformity with the ideals that inspired the jihad, they presented their policy objectives in religious terms. For instance, Usman dan Fodio's writings suggest that his concern with the establishment of a state and with the security of Muslim interests in general was connected with his belief that Muslims were obliged to reject domination by unbelievers, either by overthrowing them or by fl ight. In articulating this position and relating it to his local context, he, in explicitly Islamic ideological terms, considered his followers as jama'a and the pre-jihad Hausa rulers as unbelievers, because they had mixed Islam with “indigenous” religious practices:

The significance of the agricultural sector in the Sokoto caliphate has been a subject of interest in the past few decades. Important findings have greatly enriched our understanding of the significance of both small holdings and plantations. However, no study that exclusively deals with plantation slavery has yet comprehensively compared the importance of the non-plantation agricultural economy with that based around plantations. Such a comparison leads to two major conclusions. First, echoing Paul E. Lovejoy: unlike peasant production primarily meant for subsistence, plantations significantly shaped industrial, commercial, and other economic developments in the Sokoto caliphate; second: plantations helped to magnify state power.

It is useful to begin by relating this chapter to an earlier effort, Herbst's States and Power in Africa: Comparative Lessons in Authority and Control, published in 2000. Briefl y put, drawing on the wealth-in-people theory, Herbst argues that powerful people in Africa were not concerned with land, but rather labor, because the former was abundant while the latter was relatively scarce. Land abundance allowed African populations to be mobile, thereby making the extension of authority over neighboring people difficult, and also making it difficult to tax and maltreat laborers. Based on the challenges posed by land abundance and the “nonterritorial” nature of power in precolonial Africa, Herbst concludes, African states did not gain power through the construction of loyalties and the creation of an infrastructure meant to broadcast power, and they were weak because they exercised power solely through the use of coercion. In addition, Herbst indicates that the size of some nineteenth-century African states expanded or contracted mainly as a result of “their ability to participate in the increasing amount of international trade generated by the growing European presence on the coast of Africa” and not in response to a “war for territory.” Finally, he stresses that states that expanded in the nineteenth century did not legitimate the exercise of state power through accountability to the citizens, but by improving the technologies of coercion. Herbst's interpretation has received support from writers like Cooper.

However, Herbst's interpretation has been challenged. Austin questioned three of Herbst's ideas: that all precolonial African states were weak; that African state-makers were not concerned with land or with the construction of loyalties; and that they exercised power only through the threat of violence.

Since the 1980s several scholars have opposed the practice of examining Atlantic or world history without incorporating African history. Of these scholars, only two, Lovejoy and Enrico Dal Lago, have engaged with Dale Tomich's three related arguments: that the restructuring of the world economy which took place from the 1780s contributed to the creation and subsequent expansion of new and larger areas of slave-based agriculture in regions previously marginal to the Atlantic economy; that the second slavery of the period after 1780 was modern, in that it was characterized by expansion of production, sheer mobility and adaptability of slave labor, and the use of new techniques by planters; and that in order to properly understand the nature, establishment, and reproduction of the nineteenth-century British-centered world system one must not only focus on the new slave systems in the Americas, but also on other novel features of the Atlantic second slavery phenomenon.

Lovejoy stresses the withdrawal of Muslim/jihad states from the Atlantic world, and argues that the expansion of slavery was not only a phenomenon of the US South, Brazil, and Cuba, but also characterized the jihad states of West Africa. He seems to suggest that caliphate slavery was modern, going by the criteria outlined by Tomich. Dal Lago compares features of the slave societies in the Americas with those of the Sokoto caliphate in order to bring attention to: the unique modernity of New World planters; the common use of slave labor organized in gangs in both the caliphate and the New World; the unique form of slave resistance based on the creation of cultural forms of worship by the enslaved in the American slave societies; and the fact that religion played a major role in the assertion of the slaves’ identity in both the caliphate and the New World.

One major problem of both Dal Lago's and Lovejoy's works, however, is that they neglect East Africa. This chapter fills that gap. It seeks to stress that we should see plantation slavery in coastal East Africa as part of the Atlantic world second slavery phenomenon, and as part of a much wider global change associated with the industrial revolution.

Given that the jihad leaders emphasized Islam in their struggle against the Habe rulers and others, it is unsurprising that the application of sharia law became an important part of the sultans’ and emirs’ rule in the Sokoto caliphate; these leaders believed that at least every Muslim should be subject to the prescriptions of this law. Accordingly, they maintained many sharia courts and prohibited alcohol, prostitution, and the like. Turning to an examination of ownership and the organization of plantations, however, we see that the attitude of the sultans and emirs toward the implementation of the sharia was sometimes adaptive and non-legalistic. Although the main sources of sharia law (or the Quran, Hadith, and Sunna) do not sanction the use of slaves in large agricultural estates, rulers allowed estates of varying sizes to be attached to political offices, while at the same time allowing out-ofpower aristocrats and rich non-aristocrats to maintain private estates. There was nothing abnormal about this pattern, or in a broader sense there was nothing abnormal about Islamic rulers embracing practices or institutions not sanctioned by sharia. Indeed, most Islamic rulers since Muhammad, to varying degrees, have tended to adapt practices and institutions not sanctioned by Islamic law to Islamic use.

The establishment and maintenance of official estates was associated with an orderly pyramidal organization which linked rural administration and urban administration, albeit fostering such linkages via plantations, as through the related “Mamluk” system, was not sanctioned by Islam. Although slaves often worked on the main fields in gangs and under overseers, each estate attached to a political office was ultimately managed by the relevant political office holder who directly appropriated the surplus labor of the slaves. Private estates were similarly organized. At the same time that estate owners encouraged the punishment of recalcitrant slaves and appropriated the surplus labor of slaves, they never lost contact with the sharia, and they never failed to take into consideration the caliphate's environmental realities as well as the slaves’ humanity. They also implemented other management strategies, such as granting garden plots to slaves, which not only served to foster the notion that they were treating their slaves fairly or in accordance to the stipulations of the sharia, but that also fostered status gradation among slaves.

In a major study, Bruce S. Hall treats race as fluid and adaptable to different ideological needs and social contexts. He sees blackness in the Niger Bend as a social construction, specifically as a lack of real or concocted Arabic ancestry, and argues that those who thought of themselves as non-blacks based on this criterion often justified the enslavement of black Africans in racial terms. Although the stereotype that some Africans were black because they did not have Arabic ancestry originated outside the Sokoto caliphate, Hall argues further that the idea eventually spread throughout that state, and that arguments associated with it were deliberately used to justify enslavement in the region. In advancing his argument, Hall draws on relatively few Arabic documents. I will seek to analyze diverse source materials, including many more Arabic documents written by jihad leaders, oral sources, and secondary sources, to determine the extent to which racial theories served to justify enslavement in the Sokoto caliphate. In analyzing these sources, I will seek to bring attention to the multiple definitions of race in Hall's relatively brief discussion on the Sokoto caliphate, and in the process cast some doubt on the utility of these definitions in our understanding of jihad and enslavement. To undermine the notion that racial rhetoric was deliberately employed to secure support for the jihad, and that such rhetoric was probably used to justify enslavement in the Sokoto caliphate to a greater extent than he indicated in his study, I will seek to show that 1) blackness and a lack of real or concocted Arabic descent did not prevent many freeborn people and slaves from holding key positions in society and from enjoying related privileges; 2) that estate owners and plantation administrators mainly used ethnic and religious slurs to victimize the field laborers on a daily basis; and 3) that Sokoto caliphate royal rulers, who were mainly Fulani with concocted Arabic ancestry, embraced the pre-jihad Fulani practice of assimilating themselves into Hausa culture, and sometimes described themselves as Hausa/Hausawa.

The fact that the Fulani assimilated themselves into Hausa culture is significant, because it reveals that there was no biological basis for blackness.

This study explores the development of plantation slavery in the Sokoto caliphate from its establishment in the first decade of the nineteenth century to its demise in 1903; it demonstrates that, although favorable climatic conditions supported the development of the plantation sector in the region, political elites were also instrumental in this development. The elites embraced Islamic practices and implemented policies which aided political and economic consolidation, and promoted the expansion of agricultural production done by slaves by encouraging the immigration of potential plantation owners, increasing the availability of slaves, distributing land to wealthy individuals, and tightening security, mainly through the establishment of ribats. Merchants were also important in the expansion of the plantation sector: by creating a market for goods such as millet, sorghum, salt, cotton, and indigo both within the Sokoto caliphate and outside it, they diversified the plantation sector.

While plantations grew largely in step with political and economic developments within the caliphate and in neighboring African states, plantation profits prior to the 1860s were also derived through indirect contact with both the Atlantic world and the Mediterranean region. As the nineteenth century progressed, investments in Sokoto plantations became more directly linked with European and global capitalist markets. Plantation slavery in the Sokoto caliphate, as elsewhere in Africa, did not develop without outside stimulus, as some have suggested. Nor, as others suggest, was it shaped exclusively by the Atlantic slave trade. Rather, evidence strongly suggests that, although interaction with European and Atlantic markets was a contributing factor in the development of plantations in the Sokoto caliphate, African factors were primarily responsible for the growth of the institution.

My focus on the relationship between jihad expansion and plantation development in the Sokoto caliphate confirms that, with few exceptions, the development of plantations took place in two main phases which corresponded with the phases of the jihad expansion. Plantations were also more numerous in Kano and other emirates which were relatively prosperous and politically stable.

This book carefully avoids one pitfall manifest in some earlier studies: ignoring labor regimes. My analysis indicates that plantation slavery encompassed gang labor under constant surveillance by overseers; slaves working side-by-side with their master or directly under his supervision; and slaves working on their own account on the condition that they paid rent or such other payments to the master.