Early Mobility and Migrations

It is very likely that all human beings descend from Africans. Although the study of the evolution of humans as a species is constantly developing, and new findings could yield unexpected results, the “out of Africa” hypothesis is the most creditable one. It argues that the earliest lineages of hominids sprang from Africa and that ancestors who developed in Africa eventually supplanted variants of the species that had evolved in Europe and Eastern Asia. According to Ian Tattersall “in Africa our lineage was born, and ever since its hominids were first emancipated from the forest edges, that continent has pumped out successive waves of emigrants to all parts of the Old World.”¹ At the dawn of the existence of genus homo, the first ancestors from whom contemporary societies derive are thought to have first evolved in East Africa 1.8 to 2.5 million years ago from an earlier genus called Australopithecus.² One way of narrating human history is as a chain of migrations within and outside the African continent.

For the first millennia of human presence in Africa, lifestyles were inherently mobile. Hunting and gathering was the earliest way of producing food and required the adoption of a nomadic or semi-nomadic lifestyle. Although foraging lost its erstwhile importance in all African regions and societies, especially over the last few centuries, it never stopped co-existing alongside pastoralism and agriculture. The beginnings of animal domestication are disputed. The domestication of large cattle may have occurred as early as in the 8th millennium bce, and certainly no later than the 6th millennium bce. The domestication of grains appears to have first happened in Dynastic Egypt through the production of winter rain-fed crops. Outside the Nile Valley, grain farming is attested in the Sahel around 1800 bce and was boosted by the development of iron technology around 1000 bce. These developments are bound to have influenced migration patterns: animal husbandry induced the phenomenon of transhumance in certain regions, while farming encouraged sedentarization.

From the end of the Pleistocene period (c. 9000 bce) to about 3000 bce, greater rainfall levels accounted for the presence of Mediterranean-type vegetation and inhabitable areas in large parts of what became the Sahara desert.³ Under these favorable ecological conditions, communities living in the region could meet their subsistence requirements locally and did not have to access products available only in distant areas. Limited archaeological evidence suggests that long distance trade was not practiced until a later, drier period.⁴ The drying of the Sahara around 3000 bce engendered population redistribution that occurred over centuries. Retreat of rainforest in favor of patches of savannah in parts of West Central Africa (northern and central Cameroon, southern and central Nigeria, northern Central African Republic) facilitated the immigration of agriculturalists from the north. This process appears to have contributed to the southward spread of Bantu languages from their homeland in the Cameroon grasslands toward eastern and western axes of dispersion corresponding with the main split between West and East Bantu in West Central Africa and the Great Lakes region, respectively.⁵ Studies based on lexicostatistics advanced several hypotheses on the subsequent spread and diversification of Bantu all the way to eastern South Africa over an area of more than 11 million km².

Language spread constitutes evidence of migration. Yet reconstructing the scale, timing, direction, and magnitude of such migrations is a complex task in the absence of written sources. In a reassessment of the “Bantu expansion” debate, Jan Vansina replaced the earlier emphasis on an alleged single migration event with a model based on multiple small-scale migrations, some occurring successively and others concurrently, in various directions and each involving internal micro-movements. Language spread did not only involve migration, but also the adoption of Bantu language by autochthones in contact with Bantu-speaking immigrants, first as lingua franca and then as first language.⁶ In Bantu Studies and in Chadic Studies, theories of “Hausaization” popular in the 1960s and 1970s attributed the supposed southward spread of Chadic and proto-Hausa to mass migrations induced by climatic desiccation from the 3rd millennium bce. Since the 1990s, these interpretations have been revised in favor of hypotheses postulating multiple small-scale migrations happening over the long term, combined with language assimilation by foreign speakers.⁷ In sum, studies of language spread in Africa seem to indicate migrations of Bantu speakers from the Cameroon grasslands along various axes toward southern Africa and of Chadic speakers from around Lake Chad toward Northwest Nigeria occurring over the last 4,000 to 5,000 years. The pattern of these migrations was rhizomic rather than unidirectional, and occurred gradually through a variety of localized processes including agrarian expansion, trade and economic integration, political conquest, technological innovation, and religious conversion.

Ecological change engendered complementarities that triggered specific trade-based migratory patterns across regions. Cattle could not be kept in the more humid, tsetse fly-infested areas south of the Sahara-Sahel. Climatic complementarity encouraged exchange of cereals for animal products and the installation of regional market towns in entrepôt areas at the crossroads of major trade routes. Trade was labor intensive and involved circular migrations of relatively large groups of people. Some of the people moving along these trajectories may have been enslaved, but for regions beyond the coast and periods preceding the Muslim conquest, the evidence for early trans-regional trade in slaves is scanty. Pioneering studies of trans-Saharan migrations based on Stable Isotope Analysis (SIA) build on a combination of methods drawn from archaeometry and bioarchaeology, as well as ancient history and traditional archaeology.⁸ SIA can be used to establish whether the human remains excavated in any one site belonged to allochthonous or autochthonous individuals, and might shed light on their region of provenance. The evidence thus produced, joined with an analysis of the archaeological context for elements indicative of social status (such as burial circumstances and dietary or health profiles) may help identify slaves traded across the Sahara in the first millennium bce. But single-site findings provide only preliminary evidence on possibly broader dynamics, the magnitude and significance of which is difficult to assess.

Trade, including trade in slaves, accounted for regular migrations within and outside Africa. Trans-Mediterranean traders exploited land routes in the Maghreb in between water tracts connecting the eastern Mediterranean with European sites such as Sicily, Sardinia, and Spain. In the 9th century bce, Phoenician merchants from present-day Lebanon established a settlement in Carthage near Tunis as their base for trans-Mediterranean trade, and there were Greek settlements in Cyrenaica (in today’s eastern Libya), in the 8th century bce. Speculation abounds over the functions of Saharan chariots depicted on rock surfaces in northern Africa: should these images be seen as evidence of trans-Saharan or intra-Saharan (slave?) trade, or of hunting, or status displays? It is impossible to answer these questions with any degree of certainty, but recent studies tend to see these movements as intra-Saharan and local. Roman sources suggest a trade in black slaves, who were present in Carthaginian, Greek, and Roman towns. Were the slaves in question Saharan or Sub-Saharan in origin? Ralph Austen suggested that they might be the ancestors of Teda people from Tibesti, a highland desert zone in present-day Chad that would have been within reach of Germa, the Garamantes’ capital.⁹

In the 2nd century bce, Romans destroyed Carthage and established colonies in North Africa, where wheat and olives were cultivated for export to Roman markets. North Africa belonged to the Roman Empire, but the hinterland remained under the control of indigenous Berber populations. In late antiquity people left Africa and reached Italy, where their traces can be found in Latin, Greek, and Italian manuscripts.¹⁰ Africans seeking refuge from Vandal incursions and (from the 7th century) from the Muslim conquerors of North Africa migrated to Italy, as well as Constantinople, Spain, and Gaul.¹¹ Some of these early refugees were Christians, and their presence in other Mediterranean Christian countries, especially Italy, gave rise to devotional cults of African saints and martyrs.¹²

In antiquity, the slave trade between Africa and Europe went two-ways. Michael McCormick emphasized the importance of African markets for the trade in European—particularly Italian—slaves. He recommended caution in taking the preponderance of African markets for European captives too literally, warning that this impression may result from the prevalence of Italian sources on the early trajectories of traded slaves: Italian sources over-represented African markets because of these markets’ convenience for Italian traders.¹³ From an Africa-centered perspective this attests to intense mobility between Italy and North Africa in antiquity. Migrations of people from eastern Europe into Africa is also attested, as exemplified by the military slaves originating in the Black Sea and redeployed in Muslim Egypt for four or five centuries, until these military groups, commonly known as Mamluk, became so powerful that they claimed the Sultanate in 1250 bce.

Ancient trade in slaves and goods between North Africa and other lands possessing Mediterranean shores produced major redistributions of people. By contrast, contacts between North Africa and sub-Saharan Africa were probably less developed. Even if long-distance trade in gold, salt, grains, and slaves existed across the Sahara, it was likely limited and irregular until the introduction of the single-bump camel (Camelus Dromedarius) at the beginning of the 1st millennium ce.¹⁴ Salt and cereals were probably the main products traded across the Sahara from the Roman period until, and after, Muslim conquest.¹⁵ Desert-mined salt was also important and required by farmers in the Sudan.¹⁶ A well-established literature has been propagating the view (which some recent studies call a “myth”) that caravans brought gold to Byzantine Carthage between the 5th and 7th century ce, when Islam defeated Byzantine rule.¹⁷ Following the Arab conquest of North Africa in the 7th century and the following expansion of Islam in western and eastern Africa, trans-Saharan mobility intensified and was better documented in Arabic sources.

The spread of Islam facilitated commercial integration of regions and societies scattered across broad geographical spaces. Furthermore, Islam’s legal prohibition against enslaving Muslims created a demand for “pagan” slaves, that is, slaves captured in lands ruled by kings considered pagan. As African rulers converted to Islam, the sources of slaves who could be legitimately owned by Muslims became more distant and often harder to reach, as populations which resisted conversion retreated into less accessible areas. Muslims established their rule in Egypt in 641, and Baghdad became the capital of the Abbasid Caliphate in 762, bringing control of the Persian Gulf. Kanem-Borno around Lake Chad was Muslim by the 11th century bce. The Swahili towns of the coast of East Africa were Islamized by the 14th century, and so were some of the main West African empires: Mali and Songhai. Conversion to Islam created spaces of circulation where travelers shared the same institutions and mores across large distances. It also produced patterns of exclusion of particular groups, for whom long-distance mobility became more dangerous than in earlier times. These groups of enslavable “pagans” acquired mastery of movement over smaller-scale ecological niches: they had to know the locations of safe hiding places like mountains, forests, and caves where they could escape to when faced with the threatening raids of mounted slavers.

If the spread of Islam strengthened Muslim-controlled axes of trade and migration, it is also clear that long-distance migrations out of Africa led by non-Muslim groups continued after North Africa’s Islamization. The importance of Jewish migrants is vividly illustrated by thousands of documents, dating from the 10th to the 13th century ce, found in the Geniza of Old Cairo and studied by Shlomo Dev Goitein.¹⁸ A large section of these documents consisted of letters written by Jewish traders traveling from Tunisia and Egypt to Yemen and India.¹⁹ Goitein’s work shows that Jews in Egypt were a fully functioning diaspora. As non-Muslims they were subjected to additional poll tax (Ar: jizyah) and controls over their mobility. Jews had to carry receipts of their poll tax with them when they moved, or they could be fined if stopped by Muslim state officials.

To summarize this inevitably speculative section, in antiquity people moved in and out of the northern coast of Africa linking the continent up with other Mediterranean regions; they also circulated across the Sahara. South of the Sahara, major population movements in a southward and east-west direction (comprising internal smaller trajectories) are attested, respectively, by the spread of Bantu and Chadic languages. Until Europe started sailing to the Atlantic coast of West Africa in the 15th century, the only way for sub-Saharan African societies to come into contact with other continents was via trans-Saharan routes. These routes were crossed regularly by groups endowed with the specialist knowledge and skills required for this arduous crossing. After the 7th century, this passage included enslaved sub-Saharan Africans. Those who did not die along the way formed diasporas in the Mediterranean lands of Islam.²⁰

Africa’s Slave Trades: Forced Migrations Out of and Within Africa

Africa has been the site of some of the world’s major forced migrations of people. The trans-Saharan and Indian Ocean slave trades both preceded and outlived the better-studied Atlantic trade. Ralph Austen has been providing revised estimates of the trans-Saharan slave trade since his two seminal articles on this theme were published in 1979.²¹ In his 2010 book on trans-Saharan Africa, he argues that “between 800 and 1900 ce about 4 million people were driven across the Sahara. Another approximately 2 million came to Egypt by way of the Nile Valley from Ethiopia and the Southern Sudan, and an additional perhaps 4 million reached the Middle East and India via the Red Sea and the Indian Ocean. This total of 10 million victims is comparable to the approximately 13 million Africans forced into ships bound for the New World.”²² The trans-Saharan slave trade created diasporic communities of sub-Saharan Africans in North Africa and Islamic countries where North African Muslim trade networks operated.²³ This slow process of migration and assimilation occurred over a longer period than the Atlantic slave trade. Sub-Saharan slaves were integrated in densely populated North African regions where their phenotype stood out less than that of black slaves in the Americas; even as slaves, their status and social roles were varied. Although relations between enslaved persons and their owners were racialized in Africa, as in the New World,²⁴ the slave-master relations and racial discourses in which such relations were embedded differed in Africa and the Americas.

Slaves did not circulate across the Sahara only as chattel. They were also trade agents acting on account of their masters and were integrated in long-distance trade networks as workers looking after pack-animals, loading goods, guarding camps, cooking, and preparing shelter for temporary stops. There were many ways of being a slave, and many ways of circulating as one, ranging from the slave in fetters forced to transport him/herself to the site of his/her sale, to slaves operating as guides and travel experts.²⁵ Anjay Isa, discussed by Bruce Hall, moved at once as a commercial agent for his master’s business and to unfold his own trade projects in collaboration with other free and enslaved agents along the Ghadames-Timbuktu trade route.²⁶ The breadth of his room for maneuver depended on his master’s trust in his loyalty. Anjay’s movements were neither wholly dependent nor entirely free. Anjay’s special skills, such as his cultural and social versatility, owed him a privileged condition in spite of his slave status. Anjay’s correspondence with his master reveals his awareness of his subordinate position and fear of disappointing his patron. Yet Anjay would have had opportunities to escape, and did not take advantage of them. Escape, with the risks it entailed prior to legal abolition, would not have yielded the same rewards as loyalty.

The Indian Ocean slave trade encompassed movements of free and enslaved people between the East African coast and the South China Sea. This immense region is usually subdivided into western Indian Ocean (dominated by the Arabian Sea), and eastern Indian Ocean (dominated by the Bay of Bengal and the South China Sea). It is difficult to quantify and periodize migratory fluxes of East Africans into the Asian world, but these movements are bound to have been more important than can be surmised from limited extant sources. Slaves were traded between East Africa and China, alongside other goods, around 800 ce Much has been written about the Zanj Revolt, an important east African slave rebellion in southern Iraq in 869 ce, which resulted in slaves seizing control of cities around the Tigris River until they were defeated by the army of the Abbasid Caliphate in Baghdad in 883.²⁷ Tens of thousands of “Abyssinian” slaves are reported to have worked in the agricultural estates of the Bahrain nobility in the 10th century ce. African slaves were also traded into India.²⁸ Clearly the volume and spread of the Indian Ocean trade were of great magnitude. Yet estimates of the volume of the Indian Ocean trade are notoriously difficult to make. Sources are scattered across a large number of archives and unevenly reliable for different periods. Richard Allen cautiously gives an estimate of 565,200 as the minimum number of slaves traded by Europeans from Mozambique, the Swahili Coast, Madagascar, India, Ceylon, and Southeast Asia to destinations within the Indian Ocean basin between 1500 and 1850.²⁹ A further 449,896 to 565,193 slaves were transported from the East African coast and Madagascar to the Americas between the late 18th and the early 19th century.³⁰

The trans-Saharan slave trade remained controlled by non-European networks for its entire duration. The slaves who crossed the Sahara had been captured in Africa. By contrast, the Indian Ocean trade did not include only slaves from Africa. Until the beginning of the 1500s, it was controlled by Arab, Indian, and Chinese traders. Thereafter Portuguese, Dutch, English, and French trading companies started to compete over the control of Indian Ocean trade. These European companies operated alongside pre-existing non-European networks, some of which eventually became integrated into European operations. Africa was the most important source of slaves traded by Europeans in the Indian Ocean, with the majority of African slaves coming from Madagascar, Mozambique, and the Swahili Coast. These enslaved persons came from a broad range of African language groups, including thirteen different groups in Madagascar and over fourteen groups in East Africa, some captured in interior regions such as today’s Malawi, eastern Zambia, and Ethiopia. Slaves were also brought to the Indian Ocean from West Africa and are mentioned in the sources as Wolofs, Bambaras, and Guineans.

The trans-Atlantic trade was the largest long-distance forced migration of people in history. It lasted about 400 years, during which European and American vessels developed specifically for carrying human cargo transported over 12 million enslaved Africans to the Americas. David Eltis calculated that “cumulatively, as late as 1820, nearly four Africans had crossed the Atlantic for every European, and about four out of every five females that traversed the Atlantic were from Africa.”³¹ In the Americas, African labor enabled the development of gold and agricultural exports, and sugar plantations absorbed more than two thirds of slaves carried across the Atlantic.³² Enslaved Africans were used in a broad variety of activities in north, central, and south America, and the Caribbean. They brought with them their culture, music, religious beliefs, and technical skills. Atlantic slavery has been one of the most studied chapters of African and global history, analyzed in terms of its impacts on African societies and economies; its consequences for the Americas and Caribbean; and the internal logics of various aspects of the Atlantic trade system.

Following decades of growing abolitionist pressure, in 1807 the British abolished the slave trade. British abolition was followed by abolition in all of Europe’s major empires. These abolitions reconfigured the forced migrations of Africans out of, and within, Africa. As the Atlantic Slave Trade was abolished, so-called “legitimate trade” developed. The most important goods in this trade were groundnuts (peanuts), palm oil, and wild rubber. Rubber, cocoa, and cotton grew in importance in the early 20th century. Cocoa and palm trees grow in tropical forest zones, while peanuts grow in more arid zones but primarily along river ways: they became key cash crops in Senegambia. Cotton grows in drier savannah regions. European demand for these products in the 19th and 20th century, coupled with the sharp decrease in slave exports, induced a rise in the use of slave labor in African production and trade. Another important factor contributing to the intensification of slaving and slave trading in West Africa were a series of jihads that spread throughout the century from the Senegal River to the Cameroon highlands, waged mainly by Fulbe reformists. Wars produced slaves, some of who were fed to the trans-Atlantic trade mainly to Brazil, while others were traded in local markets. According to Paul Lovejoy, slavery became rampant in 19th century African societies.³³

Also in eastern Africa, the period between abolition and European territorial occupation saw an intensification of trade in African products. In East Africa, the British tried to obtain ivory and cloves, both in high demand in the late 18th century. This stimulated African activities for the production of cloves and the provision of ivory: cloves were planted by Omani Arabs, who used slave labor, on the East Africa Coast; ivory was obtained in the East African interior by warlords who also were major slavers. The trade of east African slaves into the Indian Ocean peaked in the mid-19th century due to a combination of factors, the most important of which was the demise of trade in the Atlantic. When in the 19th century the British increased their influence in East Africa, they did so behind the banner of abolitionism. They entered into diplomatic negotiations with local Muslim rulers in an attempt to reduce the volume of, and eventually outlaw, slave trading in this region. But the abolitionist impulse contradicted the logics of trade. As slave traders sought to avoid the West Africa Squadron patrolling Africa’s Atlantic coast, the trade in slaves in the Mozambique Channel increased fourfold:³⁴ Slave traders rounded the Cape of Good Hope to get slaves from Mozambique. The abolition of the trans-Atlantic slave trade caused a rise in the trade in slaves in eastern Africa: “the transatlantic traffic peaked in the 1780s and 1790s and went into a gradual decline about the time that the flow of slaves from the Indian Ocean began to expand from very low levels. By contrast, the Indian Ocean flow reached its own peak in the 1840s, just two decades before the Atlantic trade as a whole ended. At this point, no less than one in five of all transatlantic slaves originated in the Indian Ocean.”³⁵

In general and across regions, European acquisitions of enslaved Africans for export outside the African continent tapped into slave capturing and trading networks that continued to operate within Africa after the European abolitions of the slave trade. When looking at forced migrations, historians should consider two sets of intentionality: the ones of those forced to move and the ones of those who forced others to move. The forced migration of slaves was set in motion by economic incentives that influenced the projects of slave traders, not of enslaved persons. Yet, as soon as slaves reached their destinations they strove to acquire greater control on their lives and movements.

Migrations in the 19th Century

Following the abolition of the slave trade, European empires abolished slavery in their overseas territories. Slavery’s abolition in the British Empire in 1833, the French Empire in 1848, and in Portuguese Africa in 1869 had consequences for both intra-African and intercontinental migrations. These early 19th century abolitions happened when Europe’s territorial control over Africa was extremely limited. Therefore they affected directly only few regions where European law applied, and small numbers of Africans. In 1879, about 90 percent of the African continent still fell under African rule, and slaving and the forced migrations of enslaved persons continued unperturbed by Europe’s interference. Although the “paper partition” of Africa had been completed by the end of the 19th century, the largest part of interior Africa was territorially occupied and integrated in Europe’s administrative structures only at the beginning of the 20th century. South Africa was the main exception.

Europeans—Dutch, first, then joined by British settlers—had been present in South Africa since the end of the 17th century. Until the 18th century, low population densities had made it possible for different African groups of farmers, herders, and hunter-gatherers to coexist. These circumstances changed in the early 19th century. The consolidation of European interests in the Cape and Mozambique’s coastal regions, linking South Africa to the world economy, led to intensified demand for slaves, expansion of capitalist agriculture and mining for export, and the progressive commodification of land. These complex dynamics, combined with climatic change and forces internal to individual societies, stimulated migrations. They influenced the growing military assertiveness of the Zulu under Shaka and the displacements northwards of Sotho-Tswana speakers seeking refuge from Zulu attacks (a process referred to as mfecane, or Time of Troubles, in Zulu).³⁶ They also catalyzed the inland migrations of Boer settlers and other African groups.

Various African groups spread across space, increasing frontier conflicts with Boer farmers, who by the 1820s, had begun to face the expansion of British settlers. Following tensions over the new British abolitionist laws, British attitudes toward African groups, and land administration in the Cape’s eastern frontier, the Boers trekked out of British territory in the second half of the 1830s. The trekkers first tried to set up a Boer republic in Natal, but the latter was annexed by Britain in 1845. The trekkers continued moving northward seeking independence from British rule and control over new land.³⁷ As Norman Etherington has shown, the Boer’s trek happened alongside other “treks undertaken by different leaders, speaking different tongues, honoring different ancestors, and seeking different homelands.”³⁸ These included migrations from the Thukela River into Natal, the Rolong trek to Thaba Nchu, and the Griqua trek to Kokstad of the early 1860s.³⁹ In the 1860s, the discovery of diamond deposits in the area adjoining the intersection of the Vaal and Orange rivers created a demand for labor, which was filled primarily by African migrants. This boosted African labor migrations toward centers of mining and cash crop production that would continue to expand, governed by increasingly racist principles.

Concurrently in Western and Eastern Africa the 19th-century abolitions induced migrations of (ex-)slaves and fugitive slaves from the areas of their erstwhile enslavement toward regions of Africa under European rule, where many slaves thought they would obtain European support for their emancipation. The European administration and missionary stations supported fugitive slaves, on the whole, but help was not always forthcoming. Reverse migrations from the Americas to Africa developed, too. Implantation in the new world, initially forcible, had not erased the consciousness of belonging to Africa among many formerly enslaved Africans and their descendants in the Americas. For example, in the 1860s, during the North American civil war, several thousand African Americans migrated to Liberia.⁴⁰ Other black migrants from the Americas (especially Brazil), or re-captives liberated from slaving ships and brought back to the mainland, converted to Christianity and made it their mission to spread Christianity and abolitionist ideals in their African homelands.

In western Africa until the late 19th century, the seasonal labor migration of free workers had been a circumscribed phenomenon, confined to regions surrounding areas of intensified cash crop production.⁴¹ The participation of slaves, acting on their own account, in seasonal labor migration was limited and confined to second generation, trusted slaves, who would not take advantage of migration to escape permanently. Beyond small groups of migrant slaves, 19th century seasonal migrants were free farmers who negotiated access to farming lands with local land-owning groups.⁴² In Senegambia, free and slave migrants were temporarily employed alongside slave workers active year-round. But for many regions there are no known cases of 19^th-century antecedents of 20th-century large scale seasonal labor migrations. Travel conditions were unsafe, and the procurement of slave labor was often regulated centrally by the bureaucratic apparatus surrounding different African rulers.⁴³

Long-distance mobility was practiced primarily by workers involved in trans-regional trade, including specialist long-distance traders such as the Hausa fatake and the Manding jula in West Africa. Most of these traders were freemen, but large caravans often employed slaves in menial tasks and some specialized activities.⁴⁴ In East Africa trade in ivory boomed in the mid-19th century, and the ivory frontier regressed into the interior, lengthening the routes toward the main ports on the Zanzibar coast. African and European recruiters employed free Nyamwezi migrant workers who specialized in porterage. High demand meant that porters found work easily and were able to negotiate working conditions and regular wages.⁴⁵ Nyamwezi men left their villages for caravan porterage at the onset of the dry season in May, leaving behind women who took over household agricultural production, while migrant men worked as carriers and took on other types of migrant labor on the coast.

Free labor migration in 19th century Tanzania was important enough to warrant a generic term for the destination of labor migrations: vilimani.⁴⁶ Conditions changed with German colonization at the beginning of the 20th century. With the construction of the railway, many Nyamwezi men (and some women) gravitated toward the railway, where some of them earned wages higher than what they could have earned as porters and plantation workers. The number of migrants does not appear to have grown substantially, since it was already large, but the duration of their absence increased. This led to a restructuring of household relations and rural landscapes, as women who did not become migrants de-intensified production and cleared new fields away from chiefly control.⁴⁷ Gender (as well as status and ethnicity) influenced how people moved, and male labor migration influenced intra-household work allocation and the complementary mobility of men and women.

In the first decade of the 20th century, Europe’s occupation of almost the entire African continent and the passing of abolitionist legislation were followed by massive exoduses of slaves.⁴⁸ Given that many pre-colonial African societies were characterized by very high slave/free ratios, these spatial redistributions of slaves whose movements had been controlled and who were now able to move on their own account constituted a major population dynamic. Martin Klein estimated that slaves who returned to their regions of origin in French West Africa between 1906 and 1911 could have been over a million.⁴⁹ Paul Lovejoy and Jan Hogendorn argued that between 1897 and 1907 as many as 200,000 slaves escaped following the British conquest of the Sokoto Caliphate.⁵⁰ European administrators strove to contain these movements, but “none understood how the slaves themselves would take the issue into their own hands or, perhaps better, their own feet.”⁵¹ Migration was a central aspect in the process of emancipation.⁵²

Many ex-slaves became seasonal labor migrants, and these migrations grew in importance as a consequence of colonial rule. In his study of labor in the Jos tin mines, Bill Freund argues that the “slave population taking advantage of the end of the legal enforcement of slavery” constituted an important source of labor for the mines.⁵³ Migrants of slave and free descent had different options of social and spatial mobility. Slave descendants could either cut ties from former masters or try to take advantage of the patronage of former masters and their kinship and social networks. Some relied on the support of their masters’ diasporas abroad, as demonstrated by Lotte Pelckmans and Jean Schmitz.⁵⁴ When ex-slave migrants traveled autonomously their marginality made them vulnerable. It took time to establish host-stranger arrangements with local populations at destination, and during this time ex-slaves faced difficult conditions of life and work.⁵⁵

Poor migrants, some of slave descent, entered host societies at the bottom of labor hierarchies. Any social mobility they could achieve through their travels was hard won. Paolo Gaibazzi explored the transition between the difficult integration of immigrants into new host societies and the development of new projects of migration by the part of the migrants’ descendants in Sabi, a Soninke village in the Upper River Gambia. The host society assumed that the migrants’ ambiguous origins concealed slave descent, so they made available to them only women of slave status as brides. Accessing farming land was difficult. The descendants of immigrants were able to join profitable Europe-bound migrations later than men from allegedly local families.⁵⁶ These challenges were not unique to migrants of slave descent. As Carola Lentz has shown, all migrants in West Africa faced a range of common problems when they moved into new regions: how to define their identity and stabilize their activities in the host society; how to negotiate rights over land and other valuable assets; and how to become integrated into local societies while—if desirable—retaining a connection with their place of origin.⁵⁷

As circulation became safer under colonial rule, growing numbers of former slaves and free farmers began to migrate seasonally. Circular labor migration was encouraged by the seasonality of labor requirements in different regions. In Sahelian areas, farming was concentrated in about four months of the year, leaving eight months free for alternative occupations in tropical and coastal areas. Early forms of seasonal labor migration, such as those practiced by the laptots, the navetanes (Senegambia), and the asoba system in Ghana are documented since the end of the 18th century (in the case of the laptots)—but they appear to have been exceptional cases.⁵⁸ Circular labor migrations expanded substantially in the early 20th century.⁵⁹ People migrated in order to find the cash necessary to pay taxes for themselves and their dependents; or to avoid their fiscal obligations and forced recruitment. Progressively African colonial cities attracted rural migrants who sought access to urban jobs and services. Famously, Samir Amin saw the rise of labor migrations in this period as an outgrowth of the implantation of capitalist development: “they emigrate because the colonial system requires money of them.”⁶⁰

African Migrations in the Age of Development

In the 1930s and 1940s, the idea of development provided a new discursive legitimation for the continuation of colonial rule in eastern, western, and northern Africa.⁶¹ Concurrently, international pressure to limit labor coercion resulted in the passing of the International Labour Organization (ILO) Forced Labor Convention in 1930.⁶² The colonial state first, and independent African states afterwards, relabeled ambiguous recruitment practices as “voluntary work,” “self help,” or “human investment” and made them appear as local contributions to development. To some extent, this allowed them to circumvent the ILO’s regulations and continue recruiting cheap labor in the name of development. Migration posed a problem. Workers who could access better opportunities abroad were not available for “development” work because remuneration, if there was any, was substantially lower than their potential earnings as migrants. Measures to control the free movement of workers were put in place in most countries.⁶³

Many colonial and post-colonial development schemes involved the coercive resettlement of local groups either away from major infrastructure works, or close to worksites that required more labor than was available locally. Forced recruitment, for instance, featured prominently in the building and functioning of irrigation works in colonial West Africa’s largest agricultural development project, the Office du Niger in present-day Mali.⁶⁴ Local workers did not willingly migrate to project lands nor volunteered their work for the project’s implementation. They were coerced to move to, and work in, the project area through various strategies, including the use of the “second portion” of the annual military recruitment (a form of forced labor masked as military service). Similarly, the construction of the Cahora Bassa dam in Mozambique in the 1970s required the forced relocation of riverine communities and forced down-river farmers to abandon fertile flood plains due to the unpredictable inundations that followed the dam’s installation.⁶⁵ The Office du Niger and Cahora Bassa projects exemplify state-driven relocations in the name of development.

State efforts to manage migrations took many forms. They were aimed mainly at keeping the cost of labor low. In the interwar period, Italy’s national Committee for Emigration initially encouraged Italian migration to African colonies. But when it became clear that the cost of Italian labor was too high, a resettlement scheme was developed to move entire Somali families and communities to the irrigation schemes of the Società Agricola Italo-Somala (SAIS).⁶⁶ In the Congo, administrators who could not stop men from migrating struggled to meet their labor recruitment quotas.⁶⁷ They curtailed the migrants’ projects of self-development to make central development projects appear successful. Some administrators felt uneasy about their role. Alice Wiemers describes how District Commissioners in the Northern Provinces of Ghana controlled migration to ensure that enough labor was available in the districts for road construction and other communal labor projects. Some of them hesitated to interfere with movements that could have been productive for the colony and beneficial to migrants. As early as 1919, one District Commissioner cited by Wiemers pondered, “how one is to draw the line and distinguish when a man is a fugitive from work or when he is an emigrant to better his condition.”⁶⁸

Both colonial and independent African governments developed legal instruments to control and channel migrations.⁶⁹ In 1964, President Tubman’s vagrancy law in Liberia supported the authoritarian relocation of workers in the interest of national development plans.⁷⁰ In colonies where white settlers required local labor for farming and other businesses, control over movement took extreme racist and punitive forms. In Kenya in the 1940s and 1950s, the administration insistently decried labor shortage and the settler farmers’ inability to meet their agricultural labor requirements. Thousands of Kenyans suspected of collaborating with the Mau Mau movement were placed in concentration camps and forced to carry out public works. A Report of the Rural Wages Committee, dated 1955, highlighted that desertions from white owned farms were very high: in 1955, 10,000 workers had “illegally left employment” in sisal plantations. Raising wages as a way to retain workers on site was never considered an option, and instead the report recommended that “it should be made as difficult as possible for any worker to leave his employment in an irresponsible manner.”⁷¹ The report recommended the introduction of a punitive system of registration and movement controls that resembled South Africa’s pass laws.

The pass laws have a long genealogy in South African history. Attempts by European settlers to restrict the freedom of movement of Africans through various means go back to the early 18th century. The Native Urban Areas Act of 1923 enshrined these practices into state law by tightening and codifying requirements of African men to carry passes in South African towns and cities, declared “white” zones. Successive legal acts expanded the remit of these laws, which determined where Africans could reside, limited their ability to move freely, and made them entirely dependent on whites for their employment. Hampering mobility was tantamount to hampering freedom in a fundamental sense, because Africans who could not produce passes when stopped by the police faced arrest: “by the late 1930s, prosecutions for pass offences had peaked at some 150,000 annually, up from just over 100,000 a decade previously.”⁷² Between 1942 and 1946, the pass laws were suspended. Then at the height of apartheid in the late 1960s, the pass laws were responsible for more than 620,000 prosecutions annually.⁷³ The apartheid regime created Bantustans where black workers were forced to live and could only leave for work purposes if they held passes. Pass laws and apartheid constituted extreme forms of control over the mobility, and indeed entire life, of Africans. They developed in contexts where racist ideologies legitimized segregation and the institutionalization of white supremacy.

In most African countries the prevention or promotion of specific migratory trajectories took less coercive forms, but all states strove to channel migrations through ad hoc laws and policy measures. In the 1950s and 1960s, colonial developmentalism in western and eastern Africa went hand in hand with attempts to control a growing pool of migrant workers. Different strands of development thinking approached African migrations differently; which one prevailed determined which ideas would drive migration policy. Freddie Foks showed that the Functionalist school of British anthropology, and in particular Bronislaw Malinowski, developed a dualist model of “primitive” and “modern” economies that fit well with the vision of relations between colonial and colonized societies expounded in Frederic Lugard’s Dual Mandate.⁷⁴ This model saw “primitive” and “modern” economies as distinct spheres that could work in parallel but had to remain separate to avoid disrupting the internal functioning of each sphere. From this perspective, rural to urban migration resulting in the integration of indigenous workers in the capitalist system impacted negatively on indigenous societies, leading to their potential disintegration to the advantage of the modern sphere.

In the 1950s, the rising influence of development economists in colonial and international policy made imaginable the interaction of the “primitive” and “modern” spheres. Arthur Lewis’s 1954 influential article “Economic Development with Unlimited Supplies of Labour” recommended mobilizing labor reserves trapped into unproductive traditional institutions. The integration of rural-to-urban migrants, mine workers, and other labor migrants in capitalist business was seen as an advantageous redistribution of labor from an unproductive subsistence sector to productive capitalist enterprises. In Lewis’ view, such migrations would contribute to modernization driven by national, not colonial or foreign, initiatives. Hlya Myint’s “vent-for-surplus” model complemented Lewis’ ideas.⁷⁵ Myint focused on the potential for economic growth that resulted from “bringing into production previously unused supplies of land and labor. The advent or growth of overseas demand thus provided a market for the output of previously idle productive capacity.”⁷⁶ Generalizations about the previous idleness and unproductiveness of African labor were misguided. But these interpretations modified earlier approaches premised on the supposed incompatibility of modern and traditional economic spheres.

On the whole, these views yielded positive assessments of African labor migrations that resonated across the research-policy divide. In the mid-1960s, Elliot Berg reasoned that as long as African governments were unable to supply social services to stabilize un-skilled labor, seasonal migrant labor remunerated by piecework rather than contract would remain the most viable solution to the needs of both sending and receiving areas.⁷⁷ Later, Berg would go on to author the World Bank report “Accelerated Development in Sub-Saharan Africa: A Plan for Action.” The “Berg Report,” as it is commonly referred to, recommended that international aid organizations and African governments develop policies of economic liberalization. The Structural Adjustment Programs (SAPs) promoted by the World Bank in the 1980s and 1990s, exemplified such an approach and had mixed consequences for migration, the most obvious of which was the growth of the so-called “informal sector.”

Neoliberal development economics failed to appreciate the challenges that pushed migrants to migrate, as well as the difficult circumstances that characterized their living and working conditions as migrants. A peculiarity of African labor migrations was that the reproductive costs of labor were supported by the rural countryside from where migrants originated, and not by their capitalist employers.⁷⁸ The prevalence of return labor migration in Africa tempered the potential for proletarianization and class struggle, which in Europe followed the increasing exploitation of workers with no fallback option in the countryside.⁷⁹ African rural-to-urban migrants retained access to land by maintaining links with their families in rural areas, often sustained materially through remittances. If the continued attachment of many migrants to their region of origin was expedient to capitalist strategies, as it subsidized the social costs of labor, it was also a conscious strategy of migrants to resist full-scale dependence on wage labor.⁸⁰ Generalizations, however, should be avoided: population growth and land seizures produced rising land scarcity and the growth of African landlessness in most regions in the 20th century. In southern Africa, a history of African exclusion from land property and recent economic crisis generated a landless, unemployed class that is often described as a reserve of surplus labor that cannot be absorbed in productive businesses and therefore “migrates to depend,” that is, to make claims on relatives scattered in space.⁸¹

“Informal” Migrants, Labor Reserves, and Refugees

The bulk of regional regulation developed in western Africa in the late 1970s and 1980s emphasized the principles of free movement and freedom of residence.⁸² But African countries had, and continue to have, different employment legislation: most require foreigners to obtain resident status or work permits, or to demonstrate their capacity to support themselves financially during their stay. These requirements have costs that many migrants cannot afford. Consequently migrants often fail to comply with immigration law. They keep themselves distant from the state and its representatives because they are aware of their irregular circumstances, and because they are the targets of exactions by a corrupt police and by fraudsters who take advantage of their vulnerable position.

Migrants operating outside legal or formal employment channels are frequently described as operating in the “informal economy” or “sector.”⁸³ The notion of informality sheds little light on the positive content of the practices it claims to describe. The under-determination of informality has limited heuristic potential: “informal” migrants do not necessarily perceive their highly diverse activities as unified simply because they fall outside state regulation. State avoidance is only one aspect of some of these migrants’ strategies from their perspective. The trope of informality should be explained in relation to the political conjunctures that generated it since the 1980s, and the actual economic strategies of migrants should be explained on their own grounds, in terms of what these strategies allow migrants to achieve.

In many African countries, structural adjustment and liberalisation have increased unemployment and vulnerability. Workers who lost their jobs in the “formal” sector and did not receive adequate state support, have been forced to engage in various economic activities to meet their own and their dependents’ needs. Typically migration has been central to their livelihood strategies. Designating these strategies as “informal,” that is, falling outside the regulatory framework of the state, depoliticizes unemployment and absolves those actors who supported the implementation of liberalization policies in the first place. Furthermore, in some African countries, simple state-versus-non-state binaries conceal more than they reveal. The privatization of formerly state-owned functions has often resulted less in actual state-shrinking than in a cosmetic shift of activities from public to private businesses controlled by the same elite factions with close personal ties to the top tiers of government. Beatrice Hibou has characterized this process as “privatization of the state.”⁸⁴ The state can be an avenue to power and enrichment for particular groups, while for others it functions as an obstacle to their income generating projects. People’s migration projects are influenced by their perception of where and how they can access effective support networks. These considerations do not necessarily depend on the public or private character of institutions.

It is important to understand the functioning of networks that give access to opportunities at the top and at the bottom of African societies.⁸⁵ Which forms of economic relations enable lasting upward mobility and security, and which are nothing but the fallback options of desperately vulnerable and marginal people struggling to survive? Faced with poverty and hostile institutions, most migrants seek arrangements that fall outside state control. Employers, too, may support this state of affairs, which increases their profits and control over workers by reducing the latter’s bargaining power. In most of Africa, state avoidance is not an exceptional attitude. In many rural regions of western Africa, the majority of livelihood strategies occur outside the regulatory framework of the state. These regions remained semi-autonomous from the state and its laws during and after colonial rule.⁸⁶ But not all informal strategies are openly accessible to everyone. “Who one is”—in terms of ethnic affiliation, religion, status and gender—influences how one can move and which activities one can carry out. The primary identity that mediates participation in state-governed institutions is citizenship. But it is as members of institutions other than the nation state that many migrants are able to pursue their economic objectives. Migrants expand opportunities and diversify risk by mobilizing different identities, thereby laying claims to the support of multiple networks.⁸⁷

The four-year long trip of Tubali is a vivid illustration of migrants’ strategies. As a Nigerien Muslim Hausa speaker of Tuareg descent, Tubali could have tried to infiltrate established Tuareg and Hausa networks, but his chances in both were limited by his status. He ranked low in Tuareg hierarchical lineage structures and lacked strong connections to Hausa specialist long distance traders. His travels also differed from cin rani, smaller scale migrations aimed primarily at accessing food away from home in the dry season and carrying a connotation of social dependence. By contrast, relations among the international labor migrants (yan bida) joined by Tubali are premised on reciprocity among fellow migrants who share the same conditions of work and travel. Migrants receive help they may never return to the same individuals who assisted them, but they will provide similar support to other migrants in the future. When Tubali joined a network of labor migrants on the basis of shared religion, geographic origin, and generic ethnicity, he received money, hospitality, and help to find a job and make travel arrangements. When playing by official state rules is overly taxing, migrants embrace alternative regulatory frameworks that enable them to make claims on the resources and labor of others, not as citizens, but as kin, dependents, members of the same ethnic group, or fellow migrants from the same region.

In former settler colonies, including southern Africa, the state apparatus and its racist laws had a profound restructuring impact on precolonial African institutions. Here state avoidance per se describes more accurately the circumstances of migrants who cannot reactivate the kinds of long-established non-state rationales that govern labor relations, social hierarchies, and the right to move in other regions of Africa. In southern Africa state institutions, regulations, and bureaucratic apparatus were more developed even though they functioned for the benefit of a small section of the white population.⁸⁸ South Africa, and to a lesser extent Zimbabwe, were also characterized by levels of industrialization unparalleled in other African regions. The migrants’ experience of what the state ought to provide after the breakdown of racist regimes yielded expectations of formal employment and integration in the formal sector. These expectations, alongside the memory of segregation, apartheid, and state violence, have influenced migrant strategies.

Zimbabwe’s crisis and hyperinflation resulted in the displacement of millions of people searching for livelihoods outside the country. At the border between South Africa and Zimbabwe, migrant workers gravitate toward plantation estates, where wage labor is accessible.⁸⁹ Those offered employment in these settings become integrated, for the most part temporarily, in hierarchical structures where senior workers have relatively permanent positions and the power to facilitate the transition to more permanent employment for others. In the prevailing insecurity that defines contemporary Zimbabwe and South Africa, these farms offer highly sought-after, relatively stable livelihood options. Outside farms and similar businesses that offer the security of wages, the majority of livelihood options are precarious. But they are precarious in different ways for different categories of migrants. “Informality” as a single category conceals differences between “first-time migrants in distress,” “long-term oscillating migrant workers,” “passing-through migrants,” “short-term savers,” and others.⁹⁰

Migrants mobilize different forms of connection to enable collaboration: religious identity, ethnic solidarity, common region of origin, kinship, and belonging to bureaucratic or “traditional” hierarchies constitute strategies to produce solidarity and, consequently, to gain access to incomes in contexts marked by extreme competition over few opportunities. Borrowing from Nicholas Purcell and Peregrine Horden’s study of the Mediterranean, James McDougall and Judith Scheele referred to these strategies as “connectivity” and illustrated their functioning in Saharan contexts.⁹¹ Collaboration among migrants can intensify in response to particular political contingencies, contributing to the development of nationalism, resistance, or other political movements among diasporas.⁹² Expectations of collaboration are strongest when they are rooted in a sense of collective welfare and unity among close-knit social units, as in the case of transnational families.⁹³ They are weakest where individuals find themselves thrown together by external circumstances and display collective agency by the sheer fact of sharing marginal spaces and forming “passive networks.”⁹⁴ Encroachment upon the same space, however, does not always result in collaboration or reciprocity. It may yield competition or facilitate the intervention of “opportunity snatchers” quick to take advantage of crisis-induced changes.⁹⁵

Economic crises force businesses to restructure the labor process to remain competitive and profitable. As Tom Brass has shown, “replacing skilled, more costly and/or better organized workers already established in the capitalist labour process with less skilled, cheaper, and/or unorganized ones, enables producers to exert a downward pressure on the pay, conditions, and living standards of the proletariat generally, not just in particular national contexts.”⁹⁶ Migrant workers who do not comply with immigration law are precisely the type of vulnerable labor that finds itself tied to highly exploitative employment conditions from which they may be unable to withdraw due to debt or other forms of unfreedom. Today, these types of migrants are among the most vulnerable groups within and outside Africa. Debates on whether “informality” is good or bad for the economy, the state, or society, do little to identify the types of legal and economic measures that would improve the well being and security of those most vulnerable.

Traffickers prey on persons made vulnerable by poverty, lack of opportunities, and a deficit of belonging.⁹⁷ Although trafficking statistics are notoriously unreliable and require critical interpretation, official data on African trafficking attests to the resilience of forced migrations comparable to historical slavery. Richard Roberts and Benjamin Lawrance cite a 2006 report by the UN Office on Drugs and Crime (UNODC) that estimated that “The market for smuggling human beings from Africa to Europe in . . . transfer fees alone could be on the order of $300 million each year.” They cite Gail Wannenberg of the South African Institute for International Affairs who argued that human trafficking was the second most lucrative form of organized crime in Sub-Saharan Africa, after narcotics.⁹⁸

Refugees are yet another type of vulnerable migrants, expelled from familiar environments by conflicts and ethnic or religious persecution at home. According to the UN Refugee Agency, at the end of 2016 Africa hosted more than 5.6 million refugees, or roughly 30 percent of global refugees. A huge variety of situations define the terms by which refugees are integrated into host populations. In countries where land and agricultural resources were available and could be accessed by refugees, the latter contributed to productivity and a rise in demand of goods (especially food) sold by local societies. But often the integration of refugees into regions marked by poverty led to conflicts and environmental problems such as deforestation due to fuel wood gathering and overgrazing.⁹⁹

It must be emphasized that the distinction between refugees and economic migrants can be analytically confusing and instrumental to specific political agendas. The growing influx of African migrants into Italy and, to a lesser degree, Malta across North Africa and the Mediterranean is often referred to as a “migrant crisis” in the media. The average number of migrants traveling by boat detected reaching Italy increased from an average of 23,000 per year between 2005 and 2010 to an average of over 170,000 per year in 2014 and 2016 (with a slight decrease to 153,800 in 2015).¹⁰⁰ Responses from the EU and specific member states changed over time, but generally displayed a tendency to adopt what Jørgen Carling calls the “two kinds of people” rhetoric.¹⁰¹ Such rhetoric supports a neat categorization of those crossing the Mediterranean as either refugees deserving support and possibly asylum or economic migrants, seen as illegal and therefore detained and/or expelled.

The separation between migrants and refugees conceals the actual circumstances of African migrants willing to risk their lives in the perilous crossing (over 10,000 migrants have been reported as dead or missing in the 2014–2016 period). Many are perceived as “economic migrants” because they do not originate from countries officially recognized as “refugee-producing.” Nevertheless, their trip may have been induced by threats of violence, coercion, or extreme poverty back home. Simon McMahon and Nando Sigona highlight a second common policy misconception, namely, that the migrants’ journeys followed linear trajectories and were directed toward Europe from the start.¹⁰² In fact, research on African migrations shows that migrants often adjust their trajectories in long-term journeys that may last several years. They are responsive to changing political and economic circumstances in different transit sites, and to the conditions of relatives and other members of their social networks.¹⁰³ It is clear that refugees share many of the aspirations of economic migrants, and that many economic migrants undertook their journey under duress.

Migration has been, and continues to be, constitutive of African experiences. Much of it has been coercive and happened at an enormous human cost to migrants and their families. In the words of Achille Mbembe, “history tells us that the first thing you do to incapacitate people is to restrict their ability to move” but mobility also “allowed the stretching of societies [and] was determinant to trade and to building African civilizations.”¹⁰⁴ Migrations continue to shape the consciousness of Africans and African diasporas, as attested by the recent literature on Afropolitanism, a neologism that signals at once the specific ways in which Africans inhabit cosmopolitan identities and the ongoing vitality of debates on African migrations.

Discussion of the Literature

The study of African migrations across time and space is a fragmented field. Different historiographic debates developed in each of its sub-fields, and historians often had to adopt multi-disciplinary approaches to better understand migratory phenomena. There are also silences, especially for early historical periods—that is, themes and areas for which the historiography of African migrations is seriously underdeveloped. The work of evolutionary biologists and archaeologists dominates research on the pre-history of African migrations. The study of ancient African history is, regrettably, a narrow and specialized field. The early work of Robin Law on the Garamantes was pioneering in this respect. Michael McCormick’s monumental work on the origins of the Mediterranean economy demonstrates the importance of North Africa in this history, but there is no comparable study of cross-Mediterranean migrations from an African perspective. Research by Jonathan Conant and Stacey Graham on the circulation of people and ideas out of Africa and across the Mediterranean in late antiquity is an important opening that will hopefully continue to stimulate new research.¹⁰⁵

The study of African precolonial migrations is most developed in connection to the trans-Atlantic trade. This led to what Ned Alpers called “the tyranny of the Atlantic,” but research on trans-Saharan migrations and migrations across the Indian Ocean are rapidly growing fields.¹⁰⁶ Studies of forced migrations across the Atlantic, Indian Ocean, and Sahara have been increasingly complemented by detailed works on African diasporas at destination. Paul Lovejoy has been a driving force in this field, as he produced both single authored and (co-)edited volumes that present enslaved Africans as individuals with projects and strategies that they developed as slaves and, if relevant, after manumission.¹⁰⁷ Another well-developed field of research on precolonial migrations includes studies of the progressive Islamization of Africa and the expansion of Muslim trade and culture in Northern Africa and across the Sahara from the 7th century ce onwards. Ralph Austen’s pioneering work has been followed by research by Ghislaine Lydon, Bruce Hall, Baz Lecocq, and others focusing on more detailed, biographic, and micro-historical studies of specific migrant networks and individuals.

Since the 1950s, Africa-focused research has been trying to assess the economic efficiency of migration as a response to conditions in sending and receiving areas, or in terms of the progressive exploitation of the countryside by the logic of Capital.¹⁰⁸ Both neoliberal and dependency perspectives betrayed a functionalist interest in the causes and consequences of macro structures, but said little about the strategies of migrants. Since the 1960s, a number of detailed historical and anthropological studies have focused on the relations between migrants and host societies. In the early 1960s, the perceived threat posed by the specter of tribalism to national integration in independent African states resulted in numerous studies of the interaction of ethnicity and migration. Abner Cohen argued that ethnicity served political and economic ends.¹⁰⁹ Immanuel Wallerstein observed that, as an economic and political strategy, immigrants were forming ethno-professional communities in their new context, and that this process was a step toward the development of class consciousness.¹¹⁰ Catherine Coquery-Vidrovitch also saw the growing mass of migrant labor as “a ‘working class’ in the making.”¹¹¹ Richard Roberts and others, however, objected that the division between urban workers and rural-to-urban migrants hindered the development of class-consciousness.¹¹²

There has been a tendency to suggest that post-colonial migrations were indicative of a new social order and therefore substantially different from earlier forms of mobility. Marxist structuralist scholars like Claude Meillassoux, Jean Copans, Jean-Loup Amselle, and Pierre-Philippe Rey provided insightful interpretations of the connection between migrant labor and national and international regulatory institutions. Their studies strove to answer the question of how the articulation between capitalist and pre-capitalist modes of production occurs in Africa.¹¹³ More recently, “informality” and “informalization” have dominated debates on African migration. These approaches are often more interested in the functioning of their own models than in the question of how far these models provide accurate interpretations of the practices in question. They possibly attribute too much power to the capitalist, urban, or formal sector, as the case may be.¹¹⁴ The literature on African refugees is a relatively recent field that promises to expand as policy reports are complemented by in depth anthropological research on specific groups of African refugees within and outside Africa.