In the religious domain structural similarity often characterized belief systems within one major cultural
area. In Lower Guinea, for instance, the Gbe-speaking Aja, Fon and Ewe and the people speaking related
Yoruba dialects in the Bight of Benin (who became, respectively, the Jeje and the Nagô in Brazil) shared
the cult of ancestor monarchs and other rituals. It is, however, equally important to acknowledge the
existing religious heterogeneity within the same subzone. Various religious traditions entertained
conflictual relationships in many areas of Africa at the time of the slave trade. Even materialist conceptions
existed in some societies.^21 Islam in West Africa and Christendom in the kingdom of Kongo and in
Portuguese Angola tended to displace ‘traditional’ religions, but also absorbed many of their features,
constituting new, original fusions in Africa, long before African slaves were ‘seasoned’ in the Americas.
Because language, religion and statehood did, as a rule, not overlap, slave merchants did not ascribe the
origin of their captives—the ‘nation’ in the language of the time—in consistent ways. Slaves themselves
also adopted various identities according to the context. The complexity of African societies thus made the
construction of slave identity in the Americas an intricate process, and the changing patterns of the slave
trade and ethnic dispersal over the New World further complicated the picture. Slave traders classified their
human cargo according to their ‘nation’ because that supposedly gave buyers a better idea of the
merchandise they were acquiring. Over time particular stereotypes developed, based on European
prejudices but also on the pragmatic experiences of slave traders and plantation owners. Thus slaves from
each African ‘nation’ were attributed a set of fixed characteristics, for instance of being particularly inclined
to rebellion or, on the contrary, being very docile, hard working or lazy, inclined to practise witchcraft or
prone to commit suicide.
Captives were thus classified according to various and contradictory criteria: their port of embarkation in
Africa, the macro-region they came from, the state to which they had been subjected prior to their transatlantic
crossing, the language they spoke or the particular ethnic group they belonged to. Each system of
classification was far from consistent, however, due to the trader’s ignorance of the slaves’ specific
backgrounds, and the intricacy of African ethnic identities and political structures. For that reason
indications of slave origins in historical documents are frequently vague and often unreliable. A slave sold
as being ‘from the Guinea Coast’, for example, did not refer to any specific ethnicity or state, but referred
loosely to the West African coast.^22
Often the port of embarkation became a key marker of slave identity. Mina slaves for instance designated
those embarked from the Portuguese factory of El-Mina in present day Ghana and, by extension, other ports
of the region such as Whydah. They could be of very diverse ethnic origins. After the establishment of
Portuguese colonies in Luanda and Benguela, slaves traded from those ports were usually referred to as
‘Angolas’ and ‘Benguelas’, despite the fact that many did not come from the Portuguese colonies
themselves, but rather from neighbouring territories or states located much further inland, such as the Lunda
Empire. Thus slaves from the same ethnic background—for instance Ganguelas—could end up being
qualified as either ‘Angola’ or ‘Benguela’, according to the slave-trading network that had abducted them.
Kongo, albeit the name of a kingdom at the time, also designated a river and a wider region, so there are similar
doubts about the precise ethnic identity of ‘Congo’ slaves.^23 The Ketu, Ijesha or Ijebu nations, on the
contrary, referred to more precise West African territories and political boundaries, since they stood for some
of the states in the Bight of Benin from which slaves originated. However, slaves from these kingdoms all
spoke related dialects, and therefore could also be classified under the more general denomination Nagô,
which was used in Brazil for all speakers of what became, in Africa, the Yoruba language.^24
The indiscriminate use of these different criteria to define the slaves’ ‘nation’ makes the latter not a very
reliable category for the analysis of an original African ethnic identity. Moreover, frequent changes in the
political structure—some states expanding at the benefit of others, and eventually subjecting them—meant
THE CONTEXT OF THE BLACK ATLANTIC 35