It might be questioned whether the taking of an oath by drinking blood, as performed in the Bois Caiman, can confidently be attributed to the specific region of Dahomey, as opposed to other areas of western Africa. Ritual oaths involving the drinking of a magical potion were, certainly, not peculiar to the Dahomey area. On the Gold Coast (modern Ghana) to the west, for example, several European authors from the seventeenth century onwards describe similar practices there. Already in 1602 the Dutch trader Pieter de Marees noted that, among other forms of ritual oath, ‘some will for a further affirmation take a potion’; and the German clergyman (in Danish service) Johann Müller in the 1660s more fully describes the practice of ‘eating or drinking in the name of ... fitiso [fetish]’, though he discusses it primarily in a judicial context, as a means whereby an accused person might establish his innocence.67 Two accounts from the 1690s make explicit the rationale of the practice. The English trader Thomas Phillips notes that the potion taken ‘is to kill them the very minute that they break or violate the oath or promise they took on it’; and the Dutch trader Willem Bosman that ‘When they drink the oath-draught, ‘tis usually accompanied with an imprecation, that the fetiche may kill them if they do not perform the contents of their obligation’.68 Among later writers Rømer in the 1750s and Isert in the 1780s also describe the ritual of ‘eating fetish’ on the Gold Coast: the former, like Müller, in the context of judicial oaths, and the latter as a means of concluding military alliances between African communities.69 As in the Dahomey area, such ritual oaths were commonly used on the Gold Coast to seal agreements between Europeans and Africans. When the factors of the Royal African Company’s factory at Accra concluded an agreement with the King of Akwamu in the interior in 1681, the king’s son ‘took the fetish’ to seal the agreement; two of the Englishmen swore on the Bible, but a third took ‘their fetish’.70
Likewise in the Efik community of Old Calabar to the east (in south-eastern Nigeria), oaths were sealed by the drinking of a magical potion called mbiam, which was believed to cause anyone who swore falsely to swell up, sicken and die.71 Although not unambigously documented before the mid-nineteenth century, it is probable that the mbiam oath-draught was already used earlier, during the period of the Atlantic slave trade; a reference in the diary of the Old Calabar merchant-chief Antera Duke, in 1785, to the refusal of another chief to ‘drink doctor’, probably relates to mbiam.72 A similar ‘oath-draught’, believed to kill perjurors, was reportedly employed, apparently for judicial purposes, in Bonny.73
In these cases, however, there is no reference to the drink containing blood, whether human or animal. The Gold Coast ‘oath-draught’ is described by Müller as consisting of ‘the juice of green leaves, water and other ingedients’; by Phillips as ‘water mixed with powders of divers colours’.74Mbiam in Old Calabar was also clearly seen as distinct from the ‘blood pact’ (termed locally ‘chopping [i.e. eating] blood’, as opposed to ‘chopping doctor’, for mbiam), which appears to have come in as an innovation in the mid-nineteenth century.
The only other area of West Africa where a ritual oath-draught consisting of or including specifically blood is clearly recorded is in fact Igboland, in the hinterland of the Bight of Biafra (south-eastern Nigeria), where such ‘blood pacts’ are called igba ndu (literally, ‘binding life’).75 In the Igbo form of the ritual, the participants’ blood was smeared onto kola nuts or mixed into palm-oil to be eaten or drunk; it sometimes also included the sacrifice of chickens (or a cow), but it is not made clear whether their blood was also drunk. Here too, the ‘blood pact’ was not thought of as creating fictive kinship links, but to kill those who broke the oath. Although not clearly attested in any contemporary record before the twentieth century, there is no reason to suppose that the practice was not already established earlier, during the period of the trans-Atlantic slave trade. The Igbo form of ‘blood pact’ was used to cement commercial agreements between communities and individuals, and according to tradition played a critical role in the expansion of the commercial system of the Aro, which supplied many slaves for sale to the Europeans at the coast, during the eighteenth century.76 It was, in fact, very probably from the Igbo interior that the ‘blood pact’ was introduced into Old Calabar in the nineteenth century.
Despite the reputation of Igbo slaves in the Americas for docility (or more precisely, for expressing their disaffection through suicide rather than rebellion), it seems quite likely that it was the Igbo form of blood-oath which was utilized in some recorded slave insurrections. The conspirators in Jamaica in 1816 seem to have been predominantly Igbo, since they proposed to elect a ‘King of the Eboes’ as their leader.77 It also seems possible that the blood oath reported in Tackey’s conspiracy in Antigua in the 1730s had an Igbo origin, since Igbo were relatively numerous among the slaves taken to English Caribbean colonies (though Tackey himself was presumably, from his name, an Akan-speaker from the Gold Coast).78 It is perhaps conceivable that the Haïtian ritual oath of 1791 also derived from an Igbo rather than a Dahomian prototype, but it does not seem very likely: not only because Igbo were numerically less significant than ‘Radas’ in the Saint-Domingue slave population, but also because there is no suggestion in accounts of the rituals of the Igbo ‘blood pact’ that they involved the sacrifice of a pig.
In a recent publication, David Geggus has queried the view that the Bois Caiman ceremony was ‘simply’ a ‘Rada’ or Dahomian ritual, partly on the grounds that slaves from this region of Africa were in fact less numerous in the North than in other areas of Saint-Domingue, and suggests that its role in the insurrection in the North would be more intelligible if it was understood as ‘a syncretic bringing together of people from West and Central Africa’.79 There is clearly some force in this argument, inasmuch as ‘Congos’ formed a much greater proportion of African-born slaves in the northern plain than ‘Radas’ (over half) - although it has to be said that we do not, in fact, know precisely which slaves were involved in the Bois Caiman ceremony, or in the beginnings of the insurrection which followed it. It does not necessarily follow, however, that ‘syncretism’ is the most appropriate way of conceptualizing the putative trans-ethnic appeal of this ritual. It was arguably inherent in the nature of the ‘blood-pact’ (and of other forms of ‘oath-draught’) that they provided a means for organizing collective action independently of existing political institutions, which might therefore transcend the boundaries of existing communities, in Africa as well as in the Americas; as they served, for example, as has been seen, to cement agreements between African traders from different communities, and indeed between African and European slave-traders.
Ritual oaths, social control and slavery
Although the ‘blood pact’ clearly came to Haïti from West Africa, the social function which it served there was very different. In West Africa, ritual oaths generally functioned essentially a means of social control, reinforcing the authority of ruling elites, rather than as an organizational tactic of insurrection. On the Gold Coast, for example, Müller noted that the ceremony of ‘drinking fetish’ was employed by husbands to test and secure the fidelity of their wives.80 Likewise in Dahomey, husbands ‘drank vodun’ with their wives, and the king with his subjects, in order to secure their loyalty.81 (In the West Indies also, it will be recalled, African ritual oaths were used by husbands to prevent or detect infidelity by their wives.)
However, the practice could clearly be used against, as well as in support of, established authority. The ‘blood pact’ in Dahomey was, in fact, sometimes used for criminal purposes, by thieves or bandits.82 It could also serve as a means of organizing political opposition, as in the case of the Aplogan of Hueda in the early eighteenth century, cited earlier. The coup d’état by which King Gezo seized the Dahomian throne in 1818 likewise seems to have been based on use of the ‘blood-pact’: Gezo is said to have had over 600 ‘sworn-friends’, of whom the best known was the Brazilian slave-trader Francisco Félix de Souza, who notoriously financed the coup.83 The Kings of Dahomey, indeed, recognised the subversive potential of the blood pact, to the extent that they forbade their subjects from contracting pacts among themselves (as opposed to with the monarch, individually), since this was seen as a potential constraint on royal power.84
Such cases of the use of the ‘blood pact’ for oppositional purposes, however, related generally to factional divisions within the ruling class, rather than to the mobilization of a subordinate social group, such as insurgent slaves. In most of the few cases where ritual oaths are linked to issues concerning slavery, in fact, they served as a means of maintaining masters’ control over their slaves. On the Gold Coast in the seventeenth century, for example, it was noted that a man’s ‘newly bought slaves’, as well as his wives, were ‘bound by oath to remain faithful to their ... master’.85 Likewise in Dahomey a man would ‘drink vodun’ with his ‘servants [serviteurs]’, as well as his wives, to secure their loyalty.86
Intriguingly, the practice was adopted by at least one European slave-trader. The Englishman Thomas Shurley (who died trading off the West African coast in 1693) is said to have ‘used to make his negroes aboard take the fatish, that they would not swim ashore or run away, and then he would let them out of irons’, using for this purpose ‘a cup of English beer, with a little aloes to imbitter it’. His fellow-slaver Thomas Phillips, who reports Shurley’s practice, while acknowledging that this ‘operated upon their faith as much as if it had been made by the best fatishes in Guiney’, nevertheless observed sardonically that ‘for my part I put more dependence upon my shackles than any fatish I could give them’.87
The only parallel for a ritual oath serving in the organization of a slave insurrection in West Africa of which I am aware is the case of the society of ‘blood men’ (nka iyip)organized among slaves in Old Calabar in the nineteenth century. This was based on a blood oath, involving a mutual exchange of blood. As described by a European missionary in 1858:
They had a small quantity of blood in a plate, which they had drawn, a drop or two from those who came forward to take the oath, by tasting which, and pronouncing the oath, they entered into a covenant ... [The administrator of the oath] pulled up the wrist and cut it, drawing a drop or two of blood, which was mixed with that in the plate, and the individual took out of the blood one of the seeds, which has a symbolical significance, eat [sic: ate] it, and then dipping his fingers in the blood, put them in his mouth.88 One modern account has interpreted this ritual as establishing a form of fictive kinship,89 but the contemporary account makes clear that, as in Dahomey, it was intended rather to kill defaulters; the oath administrator ‘made a formal address to the blood, charging it to look and avenge the violation of any breach of the covenant’.90
The ‘blood pact’ in Old Calabar was clearly distinct from the mbiam oath, and no doubt represents, as suggested earlier, a borrowing from the Igbo interior - most of the slaves held in Calabar being Igbo.91 The practice is first attested in Old Calabar in the 1840s, and took on an insurrectionary character during 1850-1, when slaves on the plantations belonging to the Duke Town section of Old Calabar reportedly ‘began to bind themselves together by a covenant of blood for mutual protection’. The objective was not, strictly, as in Haïti in 1791, to overthrow slavery, but rather to resist and limit abusive treatment by their masters, and more particularly the sacrifice of slaves at chiefs’ funerals: that ‘they should not be killed for nothing, or flogged without cause’.92 The Old Calabar authorities, supported by the British Consul John Beecroft, attempted to repress the practice, negotiating a treaty with the disaffected slaves which provided that ‘no slave who has a master living shall chop blood with other slaves without special permission of the said master’, and that ‘all combinations among slaves for interfering with the correction of any domestic servant by his or her master shall be henceforth declared illegal’;93 but this attempt at control was clearly ineffective. King Eyo Honesty II, ruler of the Creek Town section of Calabar, likewise initially forbade his slaves from entering into such blood covenants with each other; but on his death in 1858, his own slaves, fearing that they might be taken for funeral sacrifices, also ‘entered into a covenant of blood to defend themselves’.94
Even in the Old Calabar case, indeed, there is an element of ambiguity, since the ‘blood pact’ was used there by masters to control their slaves, as well as by slaves to defy their masters; and indeed, even the organization of the ‘bloodmen’ was quickly co-opted into the political system. The earliest allusion to the blood oath in Old Calabar, in 1848, in fact, relates to the settlement of a dispute between a Calabar chief, Adam Duke, and his slaves, in which Duke was made to swear an mbiam oath not to punish his slaves, and at the same time ‘chopped blood’ with their headmen, as a means of reconciliation.95 By 1851 the ruler of Duke Town, King Archibong I, had made his peace with the ‘bloodmen’, and ‘joined their covenant to secure their allegiance to himself’;96 and the famous invasions of the city by the ‘bloodmen’ - in 1851, 1852 and 1871 - seem to have been more concerned with supporting rival factions among the Duke Town freemen (and especially enforcing trial by ordeal on those suspected of using witchcraft against the king) than with asserting the rights of slaves as such.97 Likewise in Creek Town, in the crisis following King Eyo Honesty II’s death in 1858, the blood oath was in the end used to secure the slaves’ allegiance to his son and successor, Young Eyo (King Eyo III). Although Young Eyo himself, as a Christian convert, declined to take any oath, the headmen of the slave communities ‘administer[ed] the oath of allegiance to all under them. They swore to be true to [Young Eyo], to hear his word and do his work; and when they deserved it, to take their punishment’, while at the same time securing a reiteration of guarantees against mistreatment: ‘they must not be killed for nothing’.98 The invasion of the town by the Creek Town ‘bloodmen’ in 1861, on the death of King Eyo, was likewise directed against the late king’s enemies rather than against oppression of slaves by masters.
That the ‘blood pact’ should have served a quite different social role in the Diaspora from in Africa should perhaps occasion no surprise, and is certainly by no means a unique case. A close parallel is provided by the case of the Cuban ‘secret society’ of Abakuá, which is clearly derived from the well-known Ekpe (Egbo) masquerade society of Old Calabar.99 Ekpe in Old Calabar was a society of wealthy merchant-chiefs (its class, rather than ethnic, character being illustrated by the fact that some European merchants were allowed to join it); and was explicitly understood to serve the function of maintaining the authority of wealthy freemen: ‘to keep women and slaves in subjection’;100 whereas Abakuá in Cuba was a society of slaves (and in the longer run, free wage-labourers) which defended their interests against their owners (or later, employers). The reality of cultural ‘continuities’ across the Atlantic does not mean that African institutions were transported unchanged, or fossilized in the Americas, any more than they were unchanging or fossilized in Africa itself.
1 Earlier versions of this paper were read at the International Roundtable on the Insurrection of the Night of 22 August 1791, Port-au-Prince, Haïti, Nov. 1997; and the Anglo-French Conference on Slavery and its Abolition, Nuffield College & Maison Française, Oxford, Dec. 1998.
29 C/Resolution 40, adopted by the General Council of Unesco, 12 Nov. 1997.
3 cf. Robin Blackburn, The Overthrow of Colonial Slavery 1776-1848 (London, 1988).
4Slavery had been abolished earlier in some individual states of what became the USA (e.g. Vermont in 1777), and declared illegal in some European countries (e.g. Scotland in 1778), but in such cases the numbers of slaves affected were very small..
5 e.g. in the classic account by C.L.R. James, The Black Jacobins: Toussaint L’Ouverture and the San Domingo Revolution (revised ed., London, 1980).
6 See e.g. Carolyn Fick, The Making of Haiti: The Saint Domingue Revolution from Below (Knoxville, 1989).
7 Haolou was a Nago (Yoruba).
8David Geggus, ‘Sugar and coffee cultivation in Saint-Domingue and the shaping of a slave labor force’, in Ira Berlin & Philip Morgan (eds), Cultivation and Culture: Work process and the shaping of Afro-American culture in the Americas (Charlottesville, Virginia, 1993), 73-98.
9 For some discussion of this issue, see John K. Thornton, ‘“I am the subject of the King of Kongo”: African political ideology and the Haitian Revolution’, Journal of World History 4/2 (1993), 181-214, esp. 199-206; David Geggus’ Slave society in the sugar plantation sones of Saint Domingue and the Revolution of 1791-3’, Slavery & Abolition, 20/2 (1999), 31-46, esp. 41.
10 Geggus, ‘Slave society’, 40-41; cf. also his earlier study, Slavery, War and Insurrection: The British Occupation of Saint-Domingue 1793-1798 (Oxford, 1982), 40-41.
11 Sydney Mintz & Richard Price, An Anthropological Approach to the Afro-American Past: A Caribbean Perpective (Philadelphia, 1976).
12 Michael Gomez, Exchanging our Country Marks: The transformation of African identities in the Colonial and Antebellum South (Chapel Hill, 1998).
13 See e.g. the classic studies by Nina Rodrigues, Os africanos no Brasil (1906, revised ed., Petropolis 1932); Fernando Ortiz, Los negros esclavos (1916; reprinted Havana, 1987).
14 Olaudah Equiano, The Interesting Narrative of the Life of Olaudah Equiano (London, 1789), ii, 101.
15See esp. Alfred Métraux, Voodoo in Haiti (trtans. Hugo Charteris, New York, 1972), 28, 86-7.
16 See,, e.g. Paul E. Lovejoy, ‘The African diaspora: revisionist interpretations of ethnicity, culture and religion under slavery’, Studies in the World History of Slavery, Abolition and Emancipation [electronic journal], 2/1 (1997); Philip D. Morgan, ‘The cultural implications of the Atlantic Slave Trade: African regional origins, New World destinations and New World developments’, Slavery & Abolition, 18 (1997), 122-45.
17 So e.g. in Cuba: Ortiz, Los negros esclavos, 45-6.
18 See further Robin Law, ‘Ethnicity and the Slave Trade: “Lucumi” and “Nago” as ethnonyms in West Africa’, History in Africa, 24 (1997), 205-19.
19 C.G.A. Oldendorp, Geschichte der Mission der evangelischen Brüder auf den caraibischen Inseln St Thomas, St Croix und St Jan (1777), translated in Soi-Daniel W. Brown, ‘>From the tongues of Africa: a partial translation of Oldendorp’s interviews’, Plantation Society, 11/1 (1983), 49. Oldendorp, in fact, also includes the ‘Nagoo’, or Yoruba, within the ‘Papaa Nation’; presumably, this sort of absorption of distinct but (in Africa) neighbouring groups reflects bilinguality among transported slaves.
20 Ortiz, Los negros esclavos, 42-3 (‘magino’, ‘sabalu’)
21See data summarized in Philip D. Curtin, The Atlantic Slave Trade: A Census (Madison, 1969), chapter 6. Also, further material in David Geggus, ‘Sex ratio, age and ethnicity in the Atlantic slave trade: data from French shipping and plantation records’, Journal of African History, 30/1 (1989), 23-44.
22Métraux, Voodoo in Haiti, 85-8. For the African oreigins of the ‘Rada’ cults, see also Guérin Montilus, Dieux en diaspora: les loa haïtiens et les vaudou du royaume d’Allada (Bénin) (Niamey, 1988). The view of Maya Deren, The Voodoo Gods (St Albans, 1975), 65-74, that the Petro cults were basically Amerindian in origin (though incorporating Congolese elements) is aberrant and implausible.
23 In Fon and other Gbe languages, vodun is the generic term for deities; whereas in Haiti it designates the system of religious practice, individual deities being termed loa.
24Cf. the suggestive (though problematic) evidence of Moreau de Saint-Méry, Description topographique, physique, civile, politique et historique de la partie française de l’île de Saint-Dominque (1797), that the Petro rite was introduced (by one Don Pedro - whence the name) only in 1768: discussed in Métraux, Voodoo in Haiti, 38-9.
25 David Geggus, ‘Sex ratio’, 36.
26In the Danish West Indies in this period, it was noted that ‘The Negroes from the Congo nation who come to the West Indies have, for the most part, a recognition of the true God and of Jesus Christ’: Oldendorp, in Brown, ‘From the tongues of Africa’, 51. For Christian continuities from Africa to the Americas more generally, cf. John Thornton, ‘On the trail of voodoo: African Christianity in Africa and the Americas’, The Americas, 44/1 (1988), 261-78; also Linda M. Heywood, ‘The Angola-Afro-Brazilian cultural connection’¸ Slavery & Abolition, 20/1 (1999), 9-23.
27John K. Thornton, ‘African soldiers in the Haitian Revolution’, Journal of Caribbean History, 25/1-2 (1991), 58-79.
28Thomas Madiou, Histoire d’Haiti (repr. Port-au-Prince, 1989), i, 96-7, 128, 131-3, 234; iii, 33.
29 For a typical statement, see James, Black Jacobins, 86: ‘Voodoo was the medium of the conspiracy’.
30Cf. David Geggus, ‘Haitian Voodoo in the eighteenth century: language, culture and resistance’, Jahrbuch für Geschichte von Staat, Wirtschaft und Gesellschaft LateinAmerikas, 28 (1991), 21-51.