CHAPTER THREE
The Clustering of African Ethnicities in the Americas
[Falupos and Arriatas are] the mortal enemies of all kinds of white men. If our ships touch their shores they plunder the goods and make the white crew their prisoners, and they sell them in those places where they normally trade for cows, goats, dogs, iron-bars and various cloths. The only thing these braves will have nothing to do with is wine from Portugal, which they believe is the blood of their own people and hence will not drink.
-Manuel Alvarez, Ethiopia Minor and a Geographical Account of the Province of Sierra Leone (c. 1615)
 

 

Despite the staggering number of Africans introduced into the Americas during the Atlantic slave trade and their crucial role in creating its wealth and forming its cultures, their origins in Africa remain obscure. There is still a widespread belief among scholars as well as the general public that Africans dragged to various places in the Americas were fractionalized and diverse, culturally and linguistically. Therefore, few of the newly arrived Africans could communicate with each other, and there was little or no basis for transmission of elements of the cultures of specific African regions and ethnicities to specific places in the Americas. This conclusion is based on anecdotal evidence as well as more complex errors in methodology. Over several generations, historians have cited statements by European and American observers at various times and places in Africa and the Americas that in order to discourage revolts, communication among new Africans was suppressed by separating and fractionalizing the various African ethnicities during their transport on Atlantic slave trade voyages as well as after they arrived in the Americas. Studies of the coastal origins of Atlantic slave trade voyages to particular places in the Americas have collapsed time, ignoring wave patterns clustering voyages originating from particular African regions, and then presented this flawed conclusion as evidence to demonstrate great diversity in the origins of enslaved Africans. Monolingual Anglophone historians have relied excessively on English-language documents and publications containing much less information about African ethnicities than documents and publications in Portuguese, Spanish, and especially French.
We now know for certain that Atlantic slave trade ships did not meander along several African coasts collecting enslaved Africans and bringing them to many different places in the Americas. Individual Atlantic slave trade ships collected Africans overwhelmingly from the same coast, usually from only one or two ports on each coast, and brought them largely to the same American port. Why? Because the longer enslaved Africans remained aboard slave trade ships, the more likely they would die before they could be sold. It is hard to believe that humanitarian concerns were a significant influence on decisions made in the Atlantic slave trade business. But spoilage of the “cargo” seriously compromised the profitability of the voyage.
If we count all the peoples of the huge African continent and the many languages they speak, we might conclude that Africans brought to the Americas were extraordinarily diverse. If we limit ourselves to the African regions from which slaves were brought in significant numbers, this diversity is substantially reduced. If we total the African coastal origins of slave trade voyages to particular regions in the Americas over several decades or centuries and collapse the span of time, we conceal the fact that Africans from the same regions and ethnicities arrived at various places in the Americas in waves.1 If we look at the changing ethnic composition of slaves exported from various African coasts over time, what we know about the patterns of the transshipment trade of Africans within the Americas, and the distribution of new Africans after their final sale, we can see further evidence of clustering of ethnicities and speakers of mutually intelligible languages on Atlantic slave trade voyages as well as after they arrived at their final destinations. We can discern the clustering of Africans from the same regions and ethnicities in local districts and on estates. This trend was almost universal: in Peru and Mexico during the sixteenth century and the first half of the seventeenth; in Brazil throughout its history; in St. Domingue/Haiti during the eighteenth century; and in mainland North American colonies and the subsequent states of the United States as well as the British West Indies during the eighteenth and nineteenth centuries. Gabriel Debien studied patterns of acquisition of new slaves on two large sugar estates over time, concluding that additional slaves were purchased from Atlantic slave trade ships in groups of ten, twenty, or thirty. “Once they were Nagos, another time Aradas, another time Ibos or Sosos, each voyage debarking one nation at a time, or else the manager preferred to take the same ethnicity from each arriving ship.”2 B. W. Higman wrote that in the British West Indies during the nineteenth century “particular source regions were more likely to predominate and a single ethnic group often accounted for a large proportion of the slaves from a particular region.”3 Aside from being clustered on estates, they were clustered in local districts. Enslaved Africans were often quite mobile and sought out their fellow countrymen living nearby.4
If we look at the bewildering variety of African ethnic designations recorded in documents in the Americas, we might come down, again, on the side of great diversity. Although a large variety of particular African ethnic designations can be identified in documents, few of them can be found with significant frequency. Thus, although Africa is a huge continent with many different peoples, only some of them were involved in the Atlantic slave trade, and relatively few African ethnicities were brought to the Americas in significant numbers.
There are multiple reasons for this clustering. Various African coasts were drawn into substantial involvement in the transatlantic slave trade in sequence over several centuries. During its early stages, European maritime trade with Africa often did not focus heavily on buying slaves to ship to the Americas. European traders named many African coasts after the major trade goods they purchased there: for example, the Gold Coast, the Pepper Coast, the Grain Coast, the Ivory Coast, the Slave Coast. During the first 150 years of the Atlantic slave trade, enslaved Africans were at first shipped mainly to Portugal, the Cape Verde Islands, the island of São Tome, or to the Gold Coast in West Africa. The Portuguese demand for slaves for labor in Africa and these Atlantic islands off the African coast impinged sharply on the number of slaves available for the transatlantic slave trade. Neither the Cape Verde Islands nor the island of Sao Tome was populated when the Portuguese first colonized them. In 1493, Portugal sent about 2,000 Jewish children under the age of eight, both male and female, to São Tome. They had been taken away from their families in Portugal and baptized before they were deported. Most of them died shortly after they arrived. Only about 6oo survived. Some of them married among themselves in the Catholic Church and had children. Most of them mated with or married Africans. Their African mates were described as very rich and intelligent. Their descendants became some of the Afro-Portuguese of Lower Guinea and West Central Africa.5
These Portuguese-colonized Atlantic islands off the African coast were launching pads for trade and colonization on the African mainland. Enslaved Africans were imported from the continent to produce very valuable trade goods. Salt, cotton, and luxurious textiles were produced in the Cape Verde Islands. They were the main products exchanged for slaves and other goods in Greater Senegambia. Throughout the centuries luxurious panos (lengths of cloth) produced in the Cape Verde Islands continued to be in very high demand on the adjoining African continent. As late as 1805, ships en route to Greater Senegambia stopped in Santiago, Cape Verde Islands, to purchase panos “greatly valued as an article of trade:”6 These Atlantic islands became the major place for exchange of domesticated plants and animals; techniques of cultivation; construction of buildings, ships, and docks; and manufacture of a variety of goods familiar in Europe, Africa, and the Americas. During the sixteenth century, the island of Sao Tome was the world’s leading sugar producer. By the 1560s, the sugar industry of Sao Tome began to be undermined by slave runaways and revolts and later by Dutch raids, invasions, and occupations.
There was relatively little demand for Africans in the Americas during the sixteenth century. Shipping technologies to the Americas were underdeveloped. Before the full impact of the demographic disasters unleashed by Spanish and Portuguese exploration, conquest, and occupation of the Americas, Native American labor was more available and certainly 00.cheaper.7 This situation changed after about 1590, when the sugar industry of Sao Tome collapsed and the Brazilian sugar industry became predominant. Between 1595 and 1640, Portugal held the asiento (contract) to supply African slaves to Spanish colonies. Cheap slaves could be obtained in Angola because of severe, extended drought and escalating warfare involving Portuguese and Dutch occupations and battles among Portuguese and Dutch allies and clients.
Once Brazil became the world’s leading sugar producer, São Tome and the Cape Verde Islands evolved into major entrepôts for the transatlantic slave trade. Africans enslaved on the continent were disembarked and refreshed with food and water, and then they worked to produce valuable trade goods while awaiting ships to take them to Portugal, to the Gold Coast, or to the Americas. Portuguese settlers of these Atlantic islands moved on to the African coast and established trading posts and settlements. They were called lançados.8Some of them were New Christians or conversos: Jews fleeing religious persecution in Portugal. Many of them moved to African communities on the mainland where religious differences were better tolerated, and they reconverted to Judaism. By 1629, there were Jewish synagogues in Recife (now Rufisque) and Cayor in Senegambia.9 With the exception of the Jewish female children sent to settle the island of São Tome in 1493, the lançados were almost entirely males. They mated with and married African women, often elite women. Their mixed-blood descendants were skilled traders, mariners, and linguists who enjoyed the great advantage of being resistant to African diseases. They played a major role in extending Portuguese trade and influence to the African continent. São Tome developed an important shipbuilding industry. The Portuguese relied heavily on the skilled mariners of São Tome and the ships built there to penetrate and conquer West Central Africa.
The lançados and their descendants developed the earliest Creole languages. These Portuguese-based Creoles were no doubt the seed for subsequent Creole languages based on French, Spanish, and English vocabularies. Cape Verdean Creole was the first Portuguese-based Creole language in Greater Senegambia. São Tome Creole was introduced into and developed in West Central Africa. The Portuguese lançados, their dependents (called grumetes), and their descendants became part of the influential Afro-Portuguese communities located along the coasts of West Africa and the riverine trade routes into the interior. They were established in enclaves, gradually influencing the surrounding areas. The Afro-Portuguese lived in places close enough to the Atlantic coast to allow for links with maritime traders. It is probably an exaggeration to describe the early generation of African slaves introduced into the Americas as a Creole generation, especially if this implies heavy European cultural influence among a significant number of Africans brought in as slaves. In coastal regions, African cultural influences based on the interpenetration of various African ethnic groups dominated the process of creolization. For example, in the coastal trading community of Luanda, Angola, 200 years after its founding by Portugal, African languages and traditional African religions continued to predominate.10 Many enslaved Africans were brought from the interior, especially after 1650, when the Dutch, English, and French transatlantic slave traders became well established.
Before 1650, the Gold Coast and the Slave Coast were an insignificant source of slaves for the Americas. The Gold Coast was a primary market for the sale of enslaved Africans within Africa. Some Africans from the Slave Coast arrived in Cartagena de Indias during the first half of the seventeenth century, but the voyages from this region were very few in number. Greater Senegambia and West Central Africa were the only significant regions of origin for the transatlantic slave trade before 1650.
Another reason that the early Atlantic slave trade focused on relatively few coasts is because it was effectively resisted in some places. During the first two centuries of the Atlantic slave trade, few enslaved Africans were collected east of Greater Senegambia/Guinea all the way to the Slave Coast. There was no significant export of slaves from Liberia, the Ivory Coast, the Gold Coast, or the Volta River basin. European ships trading along the coast between Greater Senegambia/Guinea and the Slave Coast bought gold, ivory, grain, and pepper and supplied themselves with wood, food, and water to continue their voyages. The Loango Coast between Cape Lopez Gonzalez and the mouth of the Congo/Zaire River was not involved in the early maritime slave trade. Before 1650, the Atlantic slave trade remained very sparse along these many coasts. Ivory and red-wood dyes were exported from Mayombe. The slave trade at Gabon was limited because of the defiance of its inhabitants. They had a bad reputation among the crews of the Atlantic slave trade ships. They were advised to trade from their boats and avoid going ashore unless they were well armed. Resistance to the slave trade along the Gabon Coast 11 continued over the centuries, forcing the Atlantic slave traders to concentrate on other coasts. Throughout the centuries of the Atlantic slave trade, ships trading in Senegambia and at Gabon/Cape Lopez were up to eight times more likely to experience revolts among their “cargo” than ships trading from the Slave Coast and fourteen to thirty times more likely to experience revolts than those trading at the Bight of Biafra or in West Central Africa.12
 

Africans taken as slaves in eighteenth-century Senegal. (René Claude Geoffroy de Villeneuve, L’Afrique, ou histoire, moeurs, usages et coutumes des africains: Le Sénégal, 1814. Courtesy of Special Collections, University of Virginia.)
Peoples living along other shores were difficult for the maritime slave traders to deal with. Shortly after they first arrived in Greater Senegambia, the Portuguese discovered that the Bissagos, living on islands at the mouth of the Geba River in Upper Guinea made very effective use of arrows poisoned with the spines of a fish called Bagre, which killed instantly. The Bissagos were skilled boatmen and pirates who became active kidnappers of other ethnicities to sell into the Atlantic slave trade. Near the end of the sixteenth century, the Balantas or Bagos as well as the Kru of Liberia refused to sell slaves to the Europeans. The Felupos or Diola living along the lower Casamance River in Greater Senegambia/Guinea refused to trade with the Portuguese at all.
 

Wooden collars used in the slave trade. (Thomas Clarkson, Letters on the slave-trade, and the state of the natives in those parts of Africa, ... contiguous to Fort St. Louis and Gorée, 1791. Courtesy of The Library Company of Philadelphia. )
 

A slave coffle coming from the interior in Senegal. (René Claude Geoffroy de Villeneuve, L’Afrique, ou histoire, moeurs, usages et coutumes des africains: Le Sénégal, 18l4. Courtesy of Special Collections, University of Virginia Library.)
At night, they cut the ropes anchoring the Portuguese ships, causing them to founder, and then they attacked. Many of them would not accept ransom for the Portuguese they captured and killed them instead. Resistance to the slavers continued throughout the Atlantic slave trade. At the very end of the nineteenth century, two Felupos living in Guinea-Bissau said, “We never were slaves. We never enslaved or sold our fellows.”13 There was ongoing, armed resistance to the slave traders throughout Igboland over the centuries. John N. Oriji denies “that the slave trade was a normal commercial transaction which was conducted largely in the hinterland through peaceful means. The Igbo example clearly shows that slavery and the slave trade were the primary cause of violence in the West African sub-region for over three centuries. It is also clear that without the stiff resistance mounted by many individuals and communities, slavery would have had a more devastating impact in the hinterland.” 14
As late as the 1820S, after three and a half centuries of Portuguese presence in Mozambique, the Portuguese were confined to the coast and were not allowed to enter the Makua or Yao territories. In 1857, when the slave trade involved so-called contract workers, the Makua beat off Portuguese traders trying to enter their territory and threatened to attack Portuguese coastal settlements. The Portuguese governor-general agreed not to seek “contract laborers” from their country and thereby avoided war.15
 

Poster advertising the sale of newly arrived Africans from Sierra Leone in Charleston, South Carolina, July 24,1769. (From the website “The Atlantic Slave Trade and Slave Life in the Americas: A Visual Record,” <http://hitchcock.itc.virginia.edu/Slavery>.) .)
 

Revolt aboard a slave ship, 1787. (William Fox, A Brief History of the Wesleyan Missions on the West Coast of Africa, 1851.)
Throughout West Africa, armed resistance to enslavement continued along the coasts, in the interior, along the rivers, in runaway communities, in slave pens, and aboard slave trade ships docked along the West African coasts as well as at sea.16 But the Atlantic slave traders calculated African resistance on land and sea as an inevitable cost of their lucrative trade. African resistance could at best limit it.
Until about 1650, when the Portuguese monopoly of the Atlantic slave trade was destroyed, there were only two African regions sending large numbers of enslaved Africans across the Atlantic: Greater Senegambia and West Central Africa. Africans from these two regions were clustered in the Americas. During the first half of the seventeenth century, West Central Africans were brought to Spanish North America in increasing numbers, especially to Mexico and eastern Cuba. Greater Senegambians were brought to northeastern Brazil during the sixteenth century and to far northeastern Brazil (Maranhao and Para) after 1750. But Greater Senegambians were most heavily clustered in the circum-Caribbean and Peru, and West Central Africans in Brazil.
During the seventeenth century, the English and the French began to establish trading companies operating in West Africa. After 1650, more African regions, especially the Gold Coast and the Slave Coast, became deeply involved in the Atlantic slave trade. But piracy at sea remained pivotal because of nearly unabated warfare among the European powers. Even in peacetime, large numbers of armed, unemployed seamen and former privateers were widespread throughout the Atlantic world. Estimates of the Atlantic slave trade to the Americas remain uncertain. Surviving researched documents alone tell only part of the story and give us only some of the numbers.
By the late seventeenth century, the Dutch, then the English, and then the French managed to open up the slave trade along the Loango Coast of West Central Africa north of the Congo River. This was a slow process and began mainly with the purchase of ivory. There was evidently no pool of slaves available to sell. Raids to produce slaves were eventually extended inland both north and south of the tumultuous Congo River. The Vili were the major slave raiders and sellers. During the eighteenth century, the Loango Coast became the primary source of West Central Africans in the British, French, and Dutch colonies. Substantial numbers of Kimbundu speakers from the hinterlands of Luanda, Angola, were shipped to the Loango Coast via the Malebo (Stanley) Pool. While there was significant overlap between Kikongo language group speakers shipped mainly from the Loango Coast and Kimbundu language group speakers shipped mainly from Luanda, “Angolans” in the British colonies as well as “Congo” in the French and Spanish colonies in North America were likely to be Kikongo language group speakers.
West Central Africa poses particular problems for our discussion of African ethnicities. There were conflicting usages of the terms “Congo” (Kongo) and “Angola” as coastal terms for Atlantic slave trade voyages as well as for individuals recorded in documents in the Americas. Compared to the other regions discussed, West Central Africa was the source of relatively few specific ethnicities recorded in American notarial documents in any significant numbers. A few more specific ethnic designations from West Central Africa appear in American documents as the slave trade from this region escalated during the late eighteenth and the first half of the nineteenth century. But the history, language, and culture of these peoples were so close that referring to most of them as “Kongo” is reasonable. Especially in Brazil, “Mozambique” was commonly used as a generic term for various ethnicities from that Southeast African region. More specific ethnic designations of peoples from Mozambique, mainly Makua, appear with greater frequency in French-language notarial documents.
Aside from timing, there were other factors clustering Africans from the same regions and ethnicities in the Americas. Particular African and American regions were linked by propinquity and by winds and currents affecting the length of voyages from various African coasts to various places in the Americas, as well as by the market for products sold by Atlantic slave traders along particular African coasts. Trading networks involving various types of credit arrangements, including pawnship, were established among African, Afro-European, European, and American traders.17 The North Atlantic system linked Greater Senegambia/Upper Guinea with the United States, the Caribbean, and far northeastern Brazil (Maranhão and Para). The South Atlantic system linked Central Africa, especially Angola and Mozambique, with southeastern Brazil and the Rio de la Plata (now Argentina and Uruguay). Ships leaving Portugal for Angola had to pass near northeastern Brazil. Proximity and winds minimized travel time between Angola and southeastern Brazil.18 African slavery thrived in all regions of Brazil during four centuries. Brazil was by far the greatest consumer of slaves in the Western Hemisphere.
African demand for goods sold by the Atlantic slave traders linked African and American regions. We have seen that New England rum and Virginia tobacco were very popular in the Gold Coast and Sierra Leone. The taste for tobacco was specialized. Cheap, strong, sweetened tobacco produced in Bahia, Brazil, was in great demand at Whydah and the Slave Coast.19 Neither tobacco nor gold were important imports in West Central Africa. The Loango Coast of West Central Africa was not a significant market for rum. But Brazilian rum was an essential import in Angola.
African ethnicities were clustered in the Americas because of the preference of slave owners of various regions for particular African ethnicities. There were several important reasons for these preferences. Slave owners were motivated to purchase Africans with the knowledge and skills they needed the most. The cultivation and processing of indigo, the blue dye used for \\cotton cloth, was a technology long known by Africans from Upper Guinea. During the late sixteenth century, indigo was the major item sold to the Portuguese along the Nunez River. Small quantities of indigo were used as currency. 20 In Louisiana, indigo production did not begin until Africans began to arrive from Senegal.21 During the eighteenth century, indigo became a major export crop from St. Domingue/Haiti, Louisiana, Carolina, and Central America.
The clearest example of African technology transfer to the Americas is the production of rice. Several prominent historians have argued that early slave trade voyages from Madagascar first introduced to the Americas rice and the complex technology for its cultivation. A substantial number of voyages from Madagascar to Barbados and Virginia between 1675 and 1724 have been documented. Although the British never managed to establish a permanent colony in Madagascar, they were active exporters of enslaved Africans from there, especially to Jamaica and Barbados. A census in Barbados at the end of the seventeenth century counted 32,473 slaves, half of them from Madagascar. Some of these Madagascans were surely transshipped from Barbados to Carolina, accompanied masters who were relocating there, or arrived directly on voyages of the many English and American pirates actively trading for slaves in Madagascar between 1688 and 1724. The Trans-Atlantic Slave Trade Database gives the impression that the slave trade from Madagascar was largely a British operation. Among forty-seven voyages of ships with national registry recorded in this database, forty-two (89.4 percent) were of British registry. But many early Dutch voyages as well as voyages of ships of other national registry, including Portuguese and Brazilian, were not recorded in this database, to say nothing of the many voyages of European and American pirates operating in Madagascar.22
Greater Senegambia/Upper Guinea was a major cradle of domestication and cultivation of rice as well as of many other food crops. Rice was domesticated there independently of the Asian variety.23 Generations of rice growers experimented with mini-environments, developing and adapting their techniques to varied and changing climatic conditions. Wet rice was widely cultivated using complex irrigation techniques 24 Africans from Upper Guinea were prized in Carolina and Georgia because of their skills in rice cultivation. They were less feared in the United States than in the demographically imbalanced Caribbean sugar islands, where black slaves vastly outnumbered whites. Voyages recorded in The Trans-Atlantic Slave Trade Database allow us to generalize findings to other rice-cultivating regions in the Americas as well. Although only 12.9 percent of the voyages entered in this database brought Africans from Senegambia and Sierra Leone, they were 46 percent of voyages to rice-growing regions.25 The high proportion of missing voyages from Greater Senegambia in this database allows us to tilt toward an even higher number.
Mining was another important technology transferred from Africa to the Americas. By the middle of the sixteenth century, Africans who were experienced gold miners were in demand in Colombia. Runaway slaves in Colombia were listed as “Minas” (miners). Although the meaning of the term is unclear, it changed over time and differed in various places. These “Minas” were probably experienced miners from the goldfields of Bambuk or Buré in Greater Senegambia. Africans designated as “Minas” were brought to Brazil from gold-producing regions of West Africa, including Greater Senegambia and the Gold Coast, to develop the gold-mining industry of Brazil. They were used to discover gold and develop panning and digging for it.26 Kongo were clustered in Santiago on the eastern end of Cuba to develop and work in the copper mines.27
 

Table 3.1. Transatlantic Slave Trade Voyages Bringing Enslaved Africans to Rice-Growing Regions
DestinationNumber of all VoyagesVoyages from Senegambia/ Upper Guinea
South Carolina556230(44%)
Georgia6037(62%)
Mississippi Delta3121(68%)
Florida76(86%)
Northeast Brazil8747(54%)
Total741341(46%)
While Africans with especially needed skills were favored in various parts of the Americas, specific African ethnicities were preferred for other reasons. It has become a false truism that masters were always inclined to fractionalize new Africans so they could not communicate with each other, thereby minimizing revolts among them. While this was certainly true in some cases, some masters preferred new Africans with whom they were familiar and who spoke languages understood and spoken by the slaves they already owned. There was a certain logic to bringing in Africans from “nations” who were already present in substantial numbers. The upside of creating a Tower of Babel on estates was often outweighed by the ability of partially resocialized Africans who had arrived earlier to communicate with and help resocialize newcomers. For example, in Louisiana in 1730, a master sent “un nègre de son pays” (a black from his country) to talk to a newly arrived slave whom he suspected of malingering.28 Le Page du Pratz, director of the Company of the Indies in Louisiana, who returned to France in 1732, advised Louisiana slave owners that new Africans from Guinea all believed that the French intended to kill them and drink their blood and they would kill themselves or run away shortly after they arrived unless they were reassured by the presence of older slaves from their nations 29 Moreau de St.-M6ry wrote that while some St. Domingue planters hesitated to buy Igbo slaves because of their suicidal tendencies, others preferred them because they were very attached to each other and “the newly arrived find help, care, and example from those who have come before them.” 30 A chain migration pattern has been identified for free immigrants. Those who arrived early attracted more immigrants from the same places of origin in the Old World. A modified pattern of chain migration applied to African slaves as well. Some masters preferred Africans of ethnicities who arrived early and purchased slaves from these same African “nations” when they could.
As they arrived on Atlantic slave trade voyages, new Africans were often transshipped to other regions and colonies. Patterns of this transshipment trade must be better known and understood before we can draw firm conclusions about the distribution of Africans from particular coasts and ethnicities in many places in the Americas. This is especially true for major transshipment points. With a few outstanding exceptions, very little research on this trade has been done, and little is known about it. Although at first blush the transshipment slave trade from the Caribbean seems likely to have fragmented Africans of the same regions and ethnicities because of the large numbers of ships arriving from a variety of African coasts, there was a countervailing trend indicating that preferences among both sellers and buyers tended to cluster rather than fragment arriving Africans. Some masters sent their own ships to purchase enslaved Africans from preferred coasts on Atlantic slave trade voyages as they arrived in Caribbean ports. Some of them sent their ships directly to preferred African coasts, cutting out the very expensive Caribbean middlemen. Daniel Littlefield has made very convincing links between the British Atlantic slave trade and the transshipment trade from the Caribbean. His work has revealed careful patterns in marketing that tended to cluster Africans transshipped from the British Caribbean to places where they were preferred 31 Colin Palmer has discussed preferences for African ethnicities in the transshipment trade from the British West Indies to the United States and to Spanish America between 170o and 174o. Between 1702 and 1714, before the British asiento (contract) to supply new Africans to the Spanish colonies began, at least 18,18o new Africans were transshipped from Jamaica. Fully 59.2 percent (n = 231) of British voyages bringing new Africans to Spanish American colonies between 1714 and 174o were transshipments from Jamaica.32 More recently, David Eltis has discussed the impact of preferences for Gold Coast/Slave Coast Africans in Jamaica. His conclusion is that “on Jamaican plantations at least, the estimate of two-thirds of all slaves coming from the Gold Coast-Slave Coast regions is very much a lower bound figure, with the true figure perhaps in excess of eighty percent.” 33
The preference for Gold Coast and Slave Coast Africans in Jamaica probably explains why the stated preferences of British mainland masters for Africans from the Gold Coast was not reflected in the transatlantic slave trade voyages to the United States. Ships of United States registry brought Africans they collected on the Gold Coast mainly to Jamaica and Barbados. The sharp preference for Gold Coast/Slave Coast Africans in Jamaica probably limited the transatlantic slave trade voyages from these regions to the United States and diminished their transshipment from the British Caribbean to the British mainland colonies as well. Documented Atlantic slave trade voyages from the Slave Coast to the Anglo-United States during both the colonial and the national periods are minimal.
We have little direct evidence concerning place of birth or of socialization among Africans transshipped from the Caribbean to the Anglo-United States; our knowledge about their African coastal or ethnic origins is similarly limited. But we know that they were not likely to be either born or socialized in the Caribbean. Masters were reluctant to buy Caribbean-born or socialized slaves, and for good reasons. They often had hidden illnesses, or their masters and the colonial authorities were trying to get rid of them because they were uncontrollable. Lorena S. Walsh has discerned this pattern for the Chesapeake. She argues that prestigious historians have exaggerated the number of Caribbean-born slaves brought into the British mainland colonies and states. This is certainly true for Louisiana, where the evidence is absolutely clear.34
Our newest and most systematic data about the trade in slaves shipped from Caribbean ports is from Spanish Louisiana. This colony relied very heavily on the transshipment trade rather than on the transatlantic slave trade. It is certain that almost all slaves shipped to Louisiana from the Caribbean were new Africans purchased from transatlantic slave trade voyages as they arrived in various Caribbean ports. There were both push and pull factors clustering rather than fragmenting transshipped Africans. As these voyages arrived in the Caribbean from Africa, selections among Africans from various coasts were made at the transshipment point. Africans arriving on ships coming from preferred coasts were chosen, and Africans coming from forbidden coasts were rejected. For example, a document dating from 1765 indicates that the Bight of Biafra was a forbidden coast for maritime slave traders bringing new Africans from Caribbean islands to Louisiana.35 This document explains why Peter Hill, captain of the sloop Little David, which left New York for Barbados, failed to carry out instructions to purchase between 8o and l00 newly arrived enslaved Africans to bring to the Iberville coast on the west bank of the Mississippi River opposite Baton Rouge. Captain Hill explained, “On my arrival at Barbados after doing everything in my power to fulfill the directions given me ... and finding no probability of succeeding ... I proceeded (in accordance with previous instructions) for the Island of Jamaica.... But after waiting there till the 16th of August and to that day there having but three ships arrived from Africa, two of which were of the countries excepted against and the other cargo in so bad condition that I could not Pick out the number wanted in Tolerable Order.” He could not fulfill the contract, which proved costly to his sponsors, who were successfully sued by the potential buyers of these slaves.36 Evidence from transatlantic slave trade voyages arriving in Jamaica during this time period indicates that the two ships bringing in Africans from “the countries excepted against” came from Bonny, a port on the Bight of Biafra. During this time period, such “cargoes” were probably mainly Igbos. Enslaved Africans from the Bight of Biafra were underrepresented in Spanish Louisiana, although a high proportion of voyages from the Bight of Biafra arrived in Jamaica and Cuba, both major Caribbean transshipment points for Africans brought to Louisiana.
Going beyond this one very informative, but still anecdotal document, there is significant information in the Louisiana Slave Database about slaves arriving on transshipment voyages. During the entire Spanish period we have records for 2,920 individual slaves shipped to Louisiana from the Caribbean. These figures cannot be extrapolated over time. The transatlantic slave trade as well as the transshipment trade in slaves waxed and waned with prosperity, conditions in Africa, warfare among the European powers, levels of privateering and piracy, and considerations of social control, especially after the Haitian Revolution began in 1791. The import of slaves to Louisiana was restricted or outlawed throughout the 1790S. Nevertheless, much smuggling no doubt took place.37
Among the 2,920 records describing individual slaves who arrived on ships from the Caribbean during the Spanish period (1770-1803), the origins of 967 of them were identified. Among them, 97.3 percent (n = 941) were Africans. Among these Africans, 97 percent (n = 913) were listed as brut or bozal (the French and Spanish designations for new arrivals from Africa); no other information about the origins of 87.2 percent (n = 796) of these new Africans was recorded. More specific origin information was given for 136 of them: 115 were identified by specific African ethnicities and 21 by coastal origins only. Evidence at the point of sale of these new Africans arriving on transshipment voyages from the Caribbean indicates a clustering rather than a fragmenting or randomizing of Africans of the same ethnicity. Among the slaves transshipped to Louisiana from the Caribbean, there were Africans listed as Mandingo, Kongo, and Makua. Each ethnic group was probably purchased from the same Atlantic slave trade voyage and brought over in groups on the same transshipment voyage. Many of the slaves listed under the same ethnicity were sold to the same buyer in Louisiana. Only one of the buyers, Hilario Boutte, can be identified as a jobber or reseller. In 1785, thirteen Mandingo arrived from Jamaica on the Cathalina. They were sold to four different buyers in lots of six, five, one, and one. In 1787, ten Mandingo were brought in from Martinique on the ship Nueva Orleans. They were all sold to the same buyer. Nine Kongo slaves brought in from St. Domingue in 1786 on the Rosaria were all sold to the same buyer. Ten Kongo slaves who arrived from Martinique on the Nueva Orleans in 1787 were sold to the same buyer as well. The thirty-nine Kongo slaves arriving on the Abentura from Havana in 1796 were sold to various buyers in lots ranging between eight and one. The seventeen Makua arriving from St. Domingue on the Maria Magdalena in 1785 were sold as follows: three lots of four; one lot of three, and two lots of one. The Makua ended up mainly in Pointe Coupée Parish: 54.8 percent of them (n = 23) were recorded on estates probated there.
Information about the transshipment of Africans can also be found in voyages listed in Spanish customs documents, but they are very far from complete. By 1782, slaves entered Louisiana duty-free, and the Spanish authorities were therefore not motivated to keep track of them. Except for the year 1786, information in Spanish customs documents about slaves imported into Louisiana is very sparse. The Spanish Custom House List for 1786 (see table 3.2) claimed to provide complete information, but it did not. Some voyages arriving from the Caribbean in 1786 sold slaves in Louisiana but were not listed in this document. There were 1,204 transshipped slaves sold in Louisiana in 1786 as opposed to 957 listed in the Spanish customs documents for that year.38 Nevertheless, the 1786 customs document is quite revealing about patterns of the transshipment slave trade to Louisiana from the Caribbean and has implications for other places in the Americas as well. It reveals a whole world of voyages, some of them bringing in substantial numbers of new Africans as they arrived in the Caribbean on transatlantic slave trade voyages. These transshipment voyages were not part of a large, reasonably well-documented, international network of slave trade voyages. Nor were they small, insignificant voyages either in number of voyages or numbers of slaves brought in by each shipment. For each voyage, the document includes the number of slaves brought ashore, the name of the ship, its captain and/or owner, the island of embarkation, and the date of arrival in Louisiana. It reveals that these voyages were initiated and carried out entirely by Louisiana merchants, ship captains, and slave masters, usually overlapping categories. In a large majority of instances, the captain was the owner of the ship and the sponsor of the voyage.
 

Table 3.2. Spanish Custom House List of Slaves Arriving in Louisiana from Caribbean Islands during 1786
Thus there was another world of slave trade voyages organized by slave owners who sent their own ships to the Caribbean or to Africa to collect slaves for their own use. These newly arrived Africans do not normally appear in sales documents in the Americas. If the captain was not the shipowner, he was given a few slaves from the “cargo” to sell as partial compensation for his services. Nevertheless, we find significant traces of these imported slaves in Spanish documents in Louisiana after Spain required that sellers of slaves indicate, under penalty of confiscation, how they had acquired any slave they sold 39 The master would often explain that he had brought the slaves he was selling over in his own ship, usually giving its name. The slaves almost always had been purchased in a Caribbean port.40 But sometimes the customs documents indicate an African port, such as Guinea La Cayana (a location that remains unidentified). Table 3.2 reveals that 8o percent of these enslaved Africans arrived on voyages in which the captain was also the owner.
 

Figure 3.1. Clustering of African Ethnicities in Louisiana Parishes, Spanish Period (1770-1803). Calculated from Hall, Louisiana Slave Database, 1719-1820.
The main Caribbean transshipment points for Spanish Louisiana were Jamaica, St. Dominigue, Martinique, and, after 1790, Cuba. In Louisiana documents, Africans do not at all reflect the African coastal regions of transatlantic slave trade voyages arriving in these islands during the relevant time periods. The clustering of African ethnicities during the transshipment trade from the Caribbean is evident from the heavy concentration of Africans from the Bight of Benin in Spanish Louisiana (1770-1803), especially along the Mississippi upriver from New Orleans.
The last documented transatlantic slave trade voyage from the Bight of Benin arrived in Louisiana in 1728. During the Spanish period in Louisiana (1770-1803) the transatlantic slave trade to Martinique had evidently greatly diminished. There were only twenty-three transatlantic slave trade voyages to Martinique recorded in The Trans-Atlantic Slave Trade Database between 1750 and 1795, the year the foreign slave trade to Louisiana was outlawed.41 Only three of these voyages (11.5 percent) arrived in Martinique from the Bight of Benin, and only one of them during the 178os when the transshipment trade from the Caribbean to Louisiana was most active. St. Domingue/Haiti was still importing Africans from the Bight of Benin, but the most common African coast of origin had shifted heavily to West Central Africa. Jamaican imports were predominantly from the Gold Coast, but Gold Coast slaves were extremely rare in Louisiana. The Bight of Biafra was important in the Atlantic slave trade to Cuba as well as to Jamaica during the 178os and 1790s, but relatively few Africans from this coast were found in Spanish Louisiana documents. Linking the dates of arrival of transatlantic slave trade voyages in these islands with transshipment voyages to Louisiana reveals that the Bight of Benin did not figure at all prominently as the major buying region during these years 42 We can only account for the clustering of Africans from the Bight of Benin in Louisiana through vigorous choice during the course of the transshipment trade from the Caribbean. They were heavily selected from among the voluminous voyages arriving in St. Domingue and Jamaica during the 178os when the transshipment trade to Louisiana escalated or from voyages arriving in Cuba after 1790 as the transshipment trade shifted toward that island. Africans from the Bight of Benin were present in Louisiana since the earliest years of colonization, demonstrating a significant continuity over many decades. Clustering of Africans from the Bight of Benin in upriver parishes continued through 1820.
 

Figure 3.2. Clustering of African Ethnicities in Louisiana Parishes, Early U.S. Period (1804-1820). Calculated from Hall, Louisiana Slave Database, 1719-1820.
During 1800, the foreign slave trade to Louisiana was reopened. Three transatlantic slave trade voyages were licensed by Spain and arrived in Louisiana during 1803. After Louisiana was taken over by the United States in late 1803, the foreign slave trade was immediately outlawed, but the transshipment trade from east coast ports of the United States remained legal. Once the Louisiana Territory came under United States control, the Kongo of West Central Africa became heavily clustered in Orleans and St. Charles Parishes, where the sugar industry was rapidly growing. The two documented Atlantic slave trade voyages arriving in Louisiana after the Louisiana Purchase were British ships bringing Kongo from West Central Africa. During the early U.S. period (1804-20), slaves transshipped to and sold in Louisiana arrived mainly on maritime voyages from east coast ports of the United States: Baltimore, Charleston, and Norfolk. They were overwhelmingly newly arrived Africans. Between 1804 and 1809, 63.5 percent (n = 172) of slaves arriving by sea and sold in Louisiana with recorded birthplaces were new Africans listed as brut. In 1808, thirty Kongo arrived on the transshipment voyage of the ship Ana. A surprisingly small number of slaves sold from ships arriving from east coast ports were born or were socialized in the United States. Sales documents dating from between 1810 and 1820 list 325 slaves brought from east coast ports of the United States. Very few birthplaces of slaves arriving by sea were recorded after 1810.
Many American-born slaves were no doubt brought by their masters by land, downriver, or by sea and were therefore not sold in Louisiana. Some documents record slaves who were probably sold down the Mississippi River by slave traders from Kentucky and Tennessee. We know where the slave traders lived, but not where the slaves they sold came from.
After the Louisiana Purchase in 1803, there was substantial smuggling of new Africans into the territory. The documentation for illegal voyages is, of course, thin. But young Africans of various ethnicities listed in Louisiana documents between 1804 and 1820 shed considerable light on the African ethnicities of smuggled slaves.
A comparison of the mean age of various African ethnicities recorded in Louisiana documents between 1800 and 1820 makes it abundantly clear that massive smuggling of new Africans was taking place. Very few ethnicities reflected a significant rise in mean age, which would have indicated that these Africans were elderly survivors of the legal slave trade. Young Africans were renewing the slave population. Table 3.4, calculated from the Louisiana Slave Database, is a selection of Africans of the most numerous ethnicities. This table reveals certain trends. The most significant is that there was massive smuggling of Africans into the Lower Mississippi Valley after the foreign slave trade was outlawed. It shows which ethnicities were most heavily victimized by this illegal slave trade and which were not. Judging by the substantial rise in their mean age over time, the Pular (Fulbe) and the Nard (Moor) seemed to have been the least affected. The ethnicities with dropping or nearly stable mean ages—Wolof, Kisi, Chamba, Nago, Hausa, and Mandongo as well as those categorized simply as African or of unrecognized African nation—were probably the most victimized since they were being rapidly renewed by young people. The majority of the most numerous African ethnicities show a slight increase in mean age, about two years, which would indicate that they, too, were being substantially renewed from Africa.
 

Table 3.3. Birthplace or Ethnicity of Slaves Arriving in Louisiana by Ship from East Coast Ports of the United States, 1804-1809
Source: Calculated from Hall, Louisiana Slave Database, 1719-1820.
Birthplace or EthnicityNumberPercentage
British Mainland Creole93.3
Maryland2. 7
Virginia1.4
Carolinas2.7
Native American1.4
St. Domingue31.1
Martinique1. 4
Mandingo51.8
Fulbe/Pular1. 4
Wolof62.2
Gola1. 4
Chamba Hausa1. 4
1. 4
Mina114.0
Birom1. 4
Kongo4516.5
Africa93.3
New Africans17463.5
Total274100.0
Historians have missed these numerous voyages, and no doubt similar ones initiated in other colonies, because they have emphasized centralized, mainly European archives containing records for large, commercial voyages. Most of these informal voyages were probably never documented at all. More studies of private papers and maritime documents housed in ports throughout the Americas might find traces of other such voyages. A significant number of them no doubt went directly to Africa, cutting out the expensive Caribbean middlemen and going to preferred coasts, especially to Greater Senegambia, a comparatively near destination where the slave trade was sometimes firmly in the hands of Afro-Europeans.
 

Table 3.4. Mean Age of Africans in Louisiana, 1800-1820
This chapter has argued that, although at some times and places newly arrived Africans were deliberately or randomly fragmented by the Atlantic slave trade, there were predominant countervailing patterns that tended to cluster new Africans from the same ethnicities and regions. These patterns resulted from the following factors: the tendency to load and ship enslaved Africans as quickly as possible from one coast; the gradual introduction of new regions of Africa into significant participation in the transatlantic slave trade over several centuries; geographic patterns involving distance, winds, and currents facilitating contact between specific African and American regions; traditional trading networks involving preferences for specific American products as well as long-established credit relationships on both sides of the Atlantic, including pawnship relationships in Africa;43 resistance to the Atlantic slave trade along various coasts of Africa, forcing the maritime slave traders to rely on fewer African coasts and regions; preferences for Africans from specific coasts and ethnicities, greatly influenced by African skills and technology transfer from Africa to the Americas; and preferences for Africans who were first brought into specific American regions. All of these factors established historical trade networks linking regions of Africa with regions of the Americas. Patterns in the maritime transshipment slave trade tended to cluster new Africans in places of their final destination.
The concluding four chapters of this book will link African regions and ethnicities with regions in the Americas over time.