webCôte
Library

James Steel Thayer
“The Persistence of Slavery on the Guinea Coast”

Oklahoma State University

I

To Africanists interested in the history and ethnography of the Guinea Coast, the whole issue of slavery is a complex and important one. Several significant and recent publications reflect a growing trend among scholars of Africa in general to examine or re-examine this social phenomenon in a new light—particularly Miers and Kopytoff (eds.), Slavery in Africa: Historical and Anthropological Perspectives, A. G. Hopkins, An Economic History of West Africa, Claude Meillassoux, L'esclavage en Afrique precoloniale, Frederick Cooper, Plantation Slavery on the East Coast of Africa, and Allan and Humphrey Fisher, Slavery and Muslim Society in Africa. A great many shorter works, monographs and scholarly articles have appeared in the last decade on the subject of slavery in various African societies, greatly enlarging the data and theory with which Africanists may approach or re-approach the whole subject.
The perspectives discussed in the works mentioned above have sharpened the debate about African slavery—its provenance, its nature and its function. As the volume edited by Miers and Kopytoff shows quite clearly, it is very difficult, if not altogether impossible, to discuss African slavery in terms of a single provenance, a single nature or social manifestation, or a single function throughout Africa. Questions of origin, nature and function are social and historical questions which can only be discussed and analysed in concrete social artd historical situations.
Problems of “ownership,” “rights in persons,” slaves' continuing “marginality” or degree of integration into the social system cannot be discerned or decided a priori in a continent as vast and varied as Africa. To the contrary, the multi-faceted nature of such an institution must be discovered on the basis of historical and ethnographic investigation of specific societies. The historical and cultural reality of slavery in Africa (along with other forms of servitude) is far too diverse and complex to be summarized by sweeping generalities which endeavor to encompass any and all possible instances in a single analytical framework.
Claude Meillassoux has provided an interesting typology of slavery for the latter part of the nineteenth century for West Africa. He recognizes three types of slavery in this area. First, there is “domestic slavery” in which slaves are indistinguishable from their masters. They work side by side with their masters, live in the same town, and over time their descendants are assimilated into the lineage of the masters. The slaves, introduced “… into the lineage production unit as a dependent element” (p. 64), become, over time and through his descendants, full-fledged members of his master's lineage, most often through the device of intermarriage. This type of slavery is most characteristic of the forest area of West Africa.
The second variety of slave organization was one in which the slave worked plots for his master and for himself, devoting one or two days to his own farms. In other words, the slave paid a kind of rent in labor to the master. Because of the demands of farm work, such slaves might live in the homes of their masters or in scattered hamlets near the farmland. In this form of slavery there was a distinction and separation between master and slave which prevented inter-marriage.
In the third variety of slaveholding, the number of slaves is so large that their incorporation into the larger society is difficulty (p. 64). Segregated into slave hamlets, slaves tended to form a separate social entity from thei r masters and to produce a rent in ki nd and a lesser rent in labor. If I read Meillassoux correctly, these people retain rights over their offspring and form a parallel, but related, society to that of their masters.
While the first form of slavery is typical of the forest people of West Africa, the second and third types of slavery are typical of the savannah and sahel ian regions of West Africa, at least in the nineteenth century. In a later part of the paper I will return to a discussion of Meillassoux's schema—I think that the data from the Susu and other peoples of the area will show that Meillassoux's system is not quite adequate to explain the historical development of slavery on the Guinea Coast (certainly not among the Susu). Nonetheless, it is an adequate and useful starting point.
My immediate aim in this paper is to examine the “persistence of slavery” among the Susu people and their neighbors in Sierra Leone. Although slavery is no longer practiced in the country, the “slave” among the Susu and some of their (Mulsim) neighbors is a continuing social reality. Among other peoples in and around Sierra Leone the position of slaves has tended to melt away, and this paper will also attempt to account for such differences.
By “the persistence of slavery” I do not mean to imply that there are areas in Sierra Leone where slavery is still practiced. It was abolished in 1928, and the slaves went free. However, the issues or problems of slave status, slave rights (or lack of them) and slave identity did not disappear, nor even dissipate, after emancipation. Indeed, in Sierra Leone today such issues are as strong as ever among some peoples, especially among the people among whom I conducted fieldwork for thirteen months, the Susu of Sierra Leone and Guinea.
A further question to be raised is, if slavery was at one time practically ubiquitous in Sierra Leone, why is it that some groups have retained strongly articulated ideologies, feelings, attitudes and customs relative to slaves, even to those people who are removed two or three generations from their slave ancestors, while other tribal groups not only lack such distinctions, but are even hard put to identify who, among them, is in fact of slave descent?
These questions are of interest to Africanists because, I believe, they reveal something about the nature of ideology, social systems and religion. The point will be made in this paper that the enduring ideology (and attendant customs) concerning slavery are embedded in hierarchical social systems which have,at their base, an implicit islamic ideology which contributes to the ongoing system of discrimination between freemen and slaves. The relation between religion and slavery is problematic here, involving contradition between overt and implicit ideology, but it is my contenti on that the ideology of slavery, among those who maintain rigid distinctions between “free” and “slave” is not simply due to the sediment of history, but with ideological and religious distinctions which, for these peoples, Islam calls forth and justifies.

II

Slavery has been a phenomenon well-established among peoples of the Upper Guinea Coast for a long time. Rodney maintains that slavery was unknown to Upper Guinea Coast peoples before the start of the transatlantic slave trade (Rodney:262) , but we certainly know of slavery in West African kingdoms before the inception of the slave trade—e.g., among the Dahomey and Ashanti (Fage:397). Furthermore, Martin Klein argues that the Portuguese traders who failed to note the existence of anything that we might call slavery does not imply that such an institution failed to exist “given what we know about the range of servile institutions” on the Guinea Coast (Klein:609). I would argue that many of the Muslim peoples moving into the area of the Upper Guinea Coast beginning in the fifteenth century brought with them the institution of slavery (Sanneh:82) and that Rodney's paper on the jihad of Futa Jallon shows clearly that forms of “social oppression”—to use Rodney's term—were extant in the pre-jihadic era (Rodney 1968:273).
Regardless of its exact etiology, two facts are noteworthy about slavery in the Upper Guinea Coast. First, by the end of the seventeenth century, Europeans were trading extensively for slaves. The slave trade lasted well into the nineteenth century and proved to be a significant determinant of political power and social forms throughout the entire area. Second, during the nineteenth century the slave trade was suppressed by the Europeans, and they punished those who traded in slaves and liberated captive slaves who were being tra ported on the high seas.
This much is well known. By the time of the suppression of the slave trade, according to Rodney, slaves had been amassed at various points throughout the region with no possible or safe commercial exit. Because slaves were expensive to keep, their masters put them to work on the extensive rice farms of Guinea and Sierra Leone (Rodney:1970:265-6). Although sources are clear that domestic slavery had existed before the suppression of the slave trade (and slaves had been used as items of barter), Rodney makes it clear that the vast institution of domestic slavery fully developed only when slaveowners were faced with an enormous surplus of slaves whom they could not sell, and whom they could not afford to free.
The Susu, the focus of this paper, were enthusiastic slavers (along with the Futa and Mandingo), successful not only through trading (Rodney 1970:237) but also by raiding or subduing other groups to a state of vassalage (Rodney 1970:265), a pattern that can be dated from the seventeenth century. In 1806 a famous Susu, Dala Modu, was accused by the British of slaving in the Colony of Sierra Leone (where it was illegal from the beginning) and was expelled from the Colony. It is said that Susu slavers were present near Falaba during the Sofa invasion of the 1880's to buy slaves from the victor. Throughout the nineteenth century, in war and peace, the Susu plied their trade in slaves. Governor Cardew, in his deposition in the Chalmer's report on the so-called Hut Tax War, described the following trading cycle. Susu and Fula came down from the Futa Jallon area with cattle, bound for Freetown. There they bought goods and guns. On the way upcountry they traded these items with interior peoples for slaves. Continuing to Futa Jallon, they traded the slaves for cattle. Thus the cycle repeated itself. Cardew goes on to say that the establishment of a border between Sierra Leone and Guinea may have thwarted the cycle because of duty and taxes, thus serving to divert the trade to Conakry (P.P. 1899, Pt. II, c. 8552; 44) .
The earliest accounts of Kukuna, the town on the Sierra Leone-Guinea border in which I did most of my fieldwork, reveal that the use and bartering of slaves go back at least as far as the beginning of the 19th century. Brian O'Beirne, an Irish surgeon, made a journey from Freetown to Timbo in the Futa Jallon highlands in 1821 in order to establish trade relations between the Colony of Sierra Leone and Futa Jallon. He passed through Kukuna twice, once each way on his journey, and stayed in Kukuna for about a week on each leg of his journey. In talking to a headman of Kukuna, O'Beirne notes the following in his journal:

“His horse stood near us under a shade with his fore and third feet made fast by a grass rope. I asked him (the headman) what he paid for this animal, when he answered four slaves!! I was not a little shocked at this reply, to think for a moment that an animal of the Brute creation should be valued more than four human beings. It is in the interior of this Country that the miseries of Bondage are most evident to the European. The freeman struts in his Native Robes while the harmless and unpitied Slave grovels in a state of Nature!” (Mouser: 168. The orthography and spelling are O'Beirne's).

After the formation of the Protectorate of Sierra Leone (1896), traffic in slaves was ostensibly outlawed, but the Susu, along with other peoples of the Protectorate, were permitted to keep their slaves for domestic labor. This situation prevailed until 1928 when the British government, under pressure from the League of Nations and public opinion at home, emancipated domestic slaves. This “permissive emancipation” abolished the legal status of slavery, prohibited the acquisition of new slaves and decreed that henceforth all children born to slaves were to be free.
Because of the great number of domestic slaves (see below), the colonial authorities reached the decision to emancipate the slaves unconditionally (i.e., without any compensatory payment to their owners) with some trepidation and a great deal of hesitation. In fact, the emancipation of domestic slaves turned out to be the greatest tempest in a teapot ever seen in the Colony. One advantage of the hesitation shown by the Colonial Office can be seen in the extensive surveys, reports, etc., which they undertook or commissioned before reaching their final decision. One census survey summarized here, involved a census of the slave population of Sierra Leone as of 1921. These statistics will be analysed below.

(Taken from Captain Stanley's Report of 1923, in P.P., 1928, Cmd. 3020:45).

It may be noted that all of the top categories of slaveholders are held by Muslim peoples or by peoples continguous to Muslim peoples. Stanley points out that the Wara-Wara Limba, even though far in the north and in the midst of slaveowning Muslim Susu, Mandingo and Yalunka, were nonetheless fairly isolated from these neighbors and did not own slaves, nor were they ever enslaved. By contrast, the Limba of Sela and Tonko chiefdoms were slaveowners by reason of the fact that, according to Stanley, they lived in close contact with Muslim slaveowning peoples or that a good number of them had become Muslim. (Parlimentary Paper 1928, vol. 18, cmd. 3020: 43-44). This case might be further bolstered by the example of the Vai, a Muslim people on the coastal area between Sierra Leone and Guinea, who had a great many slaves (and a high percentage of slaves) but were surrounded by non-Muslim people such as the Sherbro and Mende who had a much lower percentage of slaves.
Furthermore, the evidence at the time of emancipation showed some interesting variations between the Muslim and non-Muslim districts. In 1928 7,000 slaves in the Koinadugu District (the northernmost in Sierra Leone) left their masters—“the only place in Sierra Leone where such a 10 mass exodus occurred ” (Arkley:64) which reflects the fact that, according to Arkley, slaves in these northern, predominantly Muslim areas were treated more harshly. It is established by the Chalmer's Report that, although a provision for the self-redemption of slaves was a part of Protectorate law from its inception, there was no such provision among the Mandingo, Susu and Yalunka, nor for subsequent status mobility according to their common law. In the light of these facts, it is no wonder that so many slaves chose to emigrate.

III

I have focused this paper on the problem of slavery and slave status among one particular ethnic group, the Susu of Sierra Leone and Guinea, because I found that, even after fifty years of emancipation, there remain strong social and legal restrictions regarding slaves or, more properly put, regarding the descendants of slaves.
I became interested in the problem of African slavery during my research on problems of religion and social organization among the Susu people.
The Susu (sometimes spelled Soso or Soussou) are a prominent ethnic group in northwestern Sierra Leone and coastal Guinea. It is usually accepted that the modern Susu represent the modern descendants of Sumangoro Kante's Soso (Susu) empire, defeated by the Malinke warrior-king Sundiata around 1233 A.D. (Rodney:10). Yves Person disagrees, maintaining that the Susu were Malinke or Soninke (Serakhole), unrelated to the present-day Susu (Person: 674). Linguistically, the Susu are quite close to the Soninke, and oral traditions of the Susu support their descent from Sumangoro Kante's empire. For various reasons, political and ecological, the Susu migrated to the area of the Futa Jallon massif, establishing themselves there by the fourteenth century (Brooks: 13-14). While the upland hardwood forests of the plateau were especially attractive to Susu smiths, other Susu groups established themselves along trading routes which skirted the Futa Jallon massif. Under the influence of the Portuguese in the 16th and 17th centuries Susu traders were induced to extend trading networks towards the sea. By the eighteenth century communities of Susu were to be found from the present-day Freetown peninsula to the Rio Nunez and Rio Pongo areas of Guinea. There was a final migration under the impetus of the Fula (Peul) jihad of around 1725 under the leadership of Karamakho Alfa Ibrahim.
Many Muslim Fula had migrated with their herds into the Futa Jallon and, chafing under Susu cultural and economic domination, declared a jihad (holy war) against the predominantly pagan Susu. Those Susu who did not convert were expelled (or fled) from Futa Jallon, settling in what is known as northern Sierra Leone (and known as the Djallonke, the people of Jallon) or joining other Susu groups towards the coast (Fyfe: 3-4).
According to my ethnographic notes, there were several ways in which the Susu acquired slaves in pre-colonial times. The first, although not the foremost, was by means of war and raiding. The Susu, even though they had a strong warrior tradition, preferred trading. Slaves could be bought inland and transported by the Susu to be sold on the coast. MacCormack points out that, among the Sherbro people of Sierra Leone, trickery was sometimes used to obtain slaves for sale. One such case was reported by one of my informants who related that a Susu slave trader made a habit of travelling down the Sherbro or Gallinas Coast and arranging to marry one or two women. Upon arriving upcountry in Susu1and, he introduced his wives as “slaves” he had bought on the coast. Being far from home and relatives, the women had no one to protect them and their protestations and denials were regarded by one and all as egregious and self-serving.
According to the Susu, slavery is a social and biological concept. The fact is that slavery is an indelible mark, biologically transferable through the male line (the Susu are patrilineal). A slave man who marries a free woman has slave children, while a free man who marries a slave woman has free children. The historical accident, if you will, of being made a slave persists beyond any emancipation of slaves. Slaves are thought to be inferior a) physically, b) mentally, and c) because they are slaves. This last is by far the most important because, say my informants, it is a disgrace. It is a social disgrace, but it brings dishonor and pollutes the family biologically.
Susu domestic slaves were, as Rodney maintains, originally scattered about in small villages to work the farms of their masters. Generally all the slaves in one of these hamlets belonged to one man or one family. To this day free-born Susu may refer to a certain village as “our (or my) village,” not meaning that they were born there but that the residents of that village belonged to their family. Historically, slaves were owned outright and could be sold or inherited according to the will of their masters. As barter (cf. Lipschutz: 182) they were most often used for brideprice (two male and two female slaves being the usual sum). Absolute ownership of slaves included the right to put them to death for various offenses—murder (either of a slave or a freeman), adultery with the master's wives or female blood relatives, and for repeatedly running away.
Although slaves did try to escape (and there is a record of a revolt of Koranko slaves in Kukuna in 1838), most slaves apparently stayed where they were placed, working on the farms of their masters for five days of the week and their own farms for two. After emancipation, slaves were allowed to stay on and were given enough land for farming—this part of Sierra Leone is rather lightly.populated, and land pressure is slight.
It is difficult to gauge the economic importance of slaves in rural Susu life, but one informant (born c. 1910) told me that when he was a youth his father never worked. His “work” consisted of traveling to his slave-villages to supervise their work. Parenthetically, one wonders why Mende farmers have a reputation for diligence and hard work which the Susu lack; could it be because the Susu had slaves to do their work and thus find farm work distasteful? Lipschutz implies that the freeborn truly came to despi se manual labor and to rely for his wealth and status on the productivity of slaves (Lipschutz:193).
To this day there are distinctions between free born and “slaves” (i.e., descendants of slaves) based, first and foremost, on bloodline and descent. Although there are still social restrictions placed on slaves, many of the slaves retain a special relationship with the families of their former masters. Representatives of the free born family represent the slaves in the chiefdom at large and may do certain administrative chores in the village, such as collecting the annual head-tax from heads of household.
The social restrictions placed on slaves are very real. Slaves are precluded from running in chiefdom elections (that is, no slave may be elected paramount chief or section chief nor sit on the chief's council, although each slave village has a village headman selected from among their number). Furthermore, slaves are prohibited fpom living in “free” towns. They may not even own houses there. Presumably this is linked to the refusal of freemen to sully their bloodline through intermarriage with slaves. One informant of mine, a young man typically Susu in his social views, stated that the only person he would forbid his sister or daughter to marry was a slave—for the reasons given above. I questioned him closely about this, asking him if he would have similar objections if she were to marry a Fula or a Limba or even a Christian. Not at all. He had no objections to inter-ethnic marriage (in fact, he pointed out, one of his sisters had married a Fula) and, as for religious inter-marriage, he would have no objection if the Christian husband would permit her “to pray” (i.e., in the Muslim way).
One intersting historical piece of evidence shows that the Susu treatment of, and attitude towards, slaves has apparently altered over time. In an incident around 1838, slaves in Kukuna led by Bilali revolted and a group of them fled, establishing a slave territory in Tonko Limba. Skinner notes the following:

“According to archival sources, Lamina Bilali was a son of Alimami Nami na Sheka Dumbuya (who di ed about 1837-1838) of Kunkuna (i.e., the paramount chief of the area) and his Koranko (slave) concubine. He was raised as a free son by Alimami Namina Sheka and was promised his freedom and the freedom of his mother and siblings upon the death of the alimami. Instead, Lamina Bilali and his slave kin were treated as the property of the alimami—to be distributed as part of the inheritance. Bilali gathered his family and those slaves who wished to join him and fled in 1838 to Tonko Limba, where he established a refuge for escaped slaves and continued his activities against the Susu into the 1880s.”
(Thomas George Lawson, p. 217).

In other words, i. e., Meillassoux' s, the Susu showed a “forest” type of slavery, one in which slaves were to be incorporated over time and by means of intermarriage, into the dominant, free lineages. However, Alimami Namina Sheka's heirs and descendants had very different ideas about the place of slaves and attempted to continue to treat them as chattel and property. That is, they relegated them to the status of permanent strangers, distinguished forever by their slave status from inter-marriage with freemen. It is interesting that this revolt took place in Kukuna, implying that these slaves (and Alimami Namina Sheka's concubine) were residents in the town—as I have shown earlier, there is now a strict customary prohibition against slaves living among free Susu.
Such “confusion” is perhaps to be expected among these frontier Susu, living so close to Tonko Limba country. The Tonko Limba have certainly followed the “forest” mode of incorporation of slaves into free lineages, to the extent that it is now often impossible to discern which areas or lineages of Tonko Limba are free or slave (indeed, the current paramount chief of this area is said to be descended from slaves).
Let me cite another example to show how strong the Susu feeling about slavery is. During the fifties a young man from a slave village left the chiefdom to work in the diamond fields of Kono. After a few years of hard work he struck it rich. His behavior with his new-found wealth was exemplary— he made the pilgrimage to Mecca and the following year sent his mother on the hajj. He is still enormously wealthy—his nickname is Al-Haji Banna, banna being the Susu word for rich. When he comes to Kukuna, the capital of his chiefdom, he is received warmly by men his own age who were his childhood friends. As he sits on their verandahs, he is greeted warmly and respectfully by one and all—a true “big man” because of his wealth and his religious status. The affection and respect felt for him is genuine, and never did I hear anyone impugn his good name or character. Nonetheless, a few years ago this al-haji wanted to buy a house in Kunkuna (a “free” town) and made a very generous offer to a man with a house for sale. The freeman refused to sell it to Al-haji Banna and sold it instead to a freeman for several hundred Leones (US $1 = Leone 1.25) less than the al-haji had offered him. Neither the al-haji's money nor his status as an al-haji, nor the esteem in which he was held by everyone,could overcome the stark and ineradicable fact of his slave status.
Furthermore, to accuse a free person of being a slave is considered a term of abuse and slander. In one trial that I observed, a man had in the course of an argument accused his friend of being a slave. The friend was angry enough about this slander to press charges in the tribal court. The plaintiff was found to be in the right, and the defendant was fined 50 Leones—such a fine is no mere token. There were other such accusations made from time to time during my stay, and all of them ended up in court, with fines levied against the slanderers.
In questioning my informants about slaves, the ideology of slavery among the Susu became more unclear to me the longer I investigated. Wondering if they were considered resident outsiders, I asked them if the slaves were truly Susu. Oh, yes, they said, they are all Susu—they all speak Susu as their native tongue and, having been born in the chiefdom, they are not considered strangers. The only “foreign” trait that some of them retained was in the matter of personal names which, not being Muslim, recalled their Mende, Koranko, Kissi, or Sherbro origins. But they were considered Susu, and many of them used kinship terms with members of their “family.” An informant would tell me that so and so (a slave from “his” village) called him such and such (i.e., “elder brother,” “uncle,” “cousin,” etc.) and that he would reciprocate by calling the person in question “younger brother,” “nephew,” or “cousin,” respectively.
In the matter of religion, Susu informants also maintain that their slaves are Muslim. Indeed, every “slave” village has a small mosque.
They have their own imams and if they had few fodays or al-hajis among them, that was due to their poverty rather than any status as second-class Muslims. The example of the slave al-haji (above) puts that suspicion to rest. Theologically, the imams assured me that slaves, like anyone else, were equal in the eyes of God and that a slave who lived a righteous life would go to heaven. There is one exception to this religious egalitarianism, however. In certain categories of mosques (those larger ones referred to by the Susu as yami), slaves are not supposed to worship, especially on Friday.
If they do pray there at Friday prayers, they should pray outside. Although no one makes an issue of it if a slave should worship in the yami, most slaves would refrain from doing so because of their fear that someone would punish them by supernatural means.
In spite of this general theological equality, some of my informants told me that, in the early days of the chiefdom (probably around 1750-1800), owners of slaves had not been very anxious to islamicize their slaves. The slave owners feared, much as American slave owners feared, that religion, especially Islam, would make the slaves aware of their condition and inculcate such notions as the equality of all men or the lenient and fluid position of slaves in Muslim law. Particularly bright and clever slaves were said to have been taken by their masters to a moriman who would, by supernatural means. make the slave simple and docile. Whether this technique worked or not particularly in light of the 1838 slave revolt. We shall never know but the impulse behind such attempts is revealing.
Lastly, it should be pointed out that the Susu did not flaunt “slave” or “free” status. In slave villages I was often warned by my informants not to mention slaves—it was a touchy subject and Susu were loathe to make such social distinctions explicit. Considering the sharp and clear social distinctions which mark the difference between slave and free and the permanence of slave status, such reticence is curious for the Susu.

IV

As mentioned before, the position and treatment of slaves are not uniform throughout the Guinea Coast. They show a great deal of variation from one society to another, and in this section, I would like to emphasize some of my conclusions about slavery and slave status among the Susu by means of brief contrast with some of their neighbors on the Guinea Coast. I will summarize some of the available historical and ethnographical material and try to account for the different status of slaves in these societies.
First, as has been shown above, the position of slaves (and their descendants) among the Susu represents an incipient caste system. Castes are not unknown to Mande peoples (there is historical and ethnographical data that musicians and smiths among various Mande peoples formed endogamous castes), and “slavery seems to fit into such a social category.” Unlike social class, no amount of control of the means of production could alter the fact of slave birth and its consequences, viz., exclusion from living among the free-born and from inter-marriage with the free-born. It is clear that, for the Susu, the emancipation of slaves and their descendants in no way alters the fact of their birth. Unlike the Mende (Grace: 419) or the Sherbro (MacCormack: 198), among whom slaves could realize an “intergenerational mobility” (Miers & Kopytoff: 20)—by which slaves were in time incorporated into free-born families or at least into the wider society of free-born peoples—Susu slavery is a status and a caste that has persevered even after fifty years of emancipation. To take another example by way of contrast with the Susu, the Krim people are more lenient still. During the early part of this century. even before emancipation. a colonial official. Mr. Stanley noted that they allowed slaves to inter-marry with the free-born and to rise to positions of authority and power (Stanley in Parlimentary Papers. 1928. volume 18. cmd. 3020:43). In fact. even at this time Stanley confessed that he often confused slave-born and free-born among the Krim—one Krim an explained that certain people in the village (second and third generation slaves) were easily confused with the free-born because “we have never told them they are slaves.” The difference between the Krim and the Susu could hardly be greater—among the Susu, not only would slaves know they were slaves, but they would be segregated in life and marriage. There would be no inter-marriage (faithful to the caste-like principle of endogamy) and slave-born people would be categorically excluded from positions of power and authority.
Svend Holsoe points out that slaves among the (Muslim) Vai came to occupy positions of political authority and power. However, this was illusory.
A slave might be appointed the official town chief in order to deal with far-off Liberian officials. but “… power still resides with the eldest member of the founding lineage, and all important decisions must be made by him in conjunction with the [free-born] elders of the town.” (Holsoe: 300). The Susu permitted slave villages to have their own (slave) headmen, but the Susu would never permit slaves to have any political authority, even like the symbolic power of the Vai slaves, for slaves by definition may never have authority or precedence over the free-born. MacCormack identifies a situation among the Sherbro in which the adopted “son” (i .e., slave) of a certain freeman was even made paramount chief. Such an event would be impossible among the caste-conscious Susu—part of the process of the nomination of candidates for the office of paramount chief involves careful scrutiny of the candidates' bloodlines by the deceased paramount chief's council in order to determine, among other things, whether the candidate is of free or slave origin. If he is of slave descent, he is forbidden from standing for election.
A second point to note is that the Susu maintain this system of slave caste, as they maintained it in the past, in part by the fact of physical separation from slaves and the lack of economic resources owned or controlled by slaves. Slaves were isolated in villages where they worked the farms of their masters (this at least has been the pattern known to us since White contact). Susu slaves were (and their descendants still are) separated from free towns, confined to villages in the remoter areas of their chiefdoms. The opportunities for mingling with freemen are limited; this also limits the opportunities or occasions for inter-marriage. Thus, by physical separation and cultural prohibition Susu slaves are kept as non-relatives, as permanent strangers. Because of this physical isolation, it is also more difficult for them to engage in trade. Their villages are often off the beaten track, and trading is difficult because of the distance and problems of 'transportation. Slaves are effectively denied access to the one activity in which the Susu engage that is able to advance them economically. It is also significant because, as is well-known to historians of Africa, trade is the means whereby strangers come to settle among foreign peoples and, because of their wealth and prestige, to intermarry with the local people. Physical separation and lack of economic opportunities, both singly and together, work to perpetuate the Susu slave caste system.
Vaughan has characterized slavery as among the Margi as a 'limbic' institution. The marginality of slaves has been formalized, “for its members exist in the hem of society, in a limbo, neither enfranchised … nor true aliens.” (100) This pattern, typical of the Susu, is curiously absent among many of their neighbors. The Mende and Sherbro pattern of slave incorporation seems to follow the African pattern in which, after a period of generational time, these strangers (albeit unwilling ones) were drawn into the host population through adoption and inter-marriage. This, indeed, is the pattern which permitted the diffusion of Islam through much of this region (see Hopewell, Skinner, Howard & Willis). Muslim traders settled along trade routes and entered into client relations with the indigenous chief and, due to his wealth and learning soon married into one of the prominent families of the chiefdom. Any ethnography of the peoples of Sierra Leone can show the traces of Mandinka, Koranko, Susu, Yalunka or Fula “strangers” whose progeny have long since become “citizens” of the area, indistinguishable (at least in any pejorative sense) from the native inhabitants among whom'they live.
Such does not appear to be the situation among the Muslim peoples (particularly the Susu/Yalunka, Fula and Mandinka peoples) in their homeland with regard to their slaves. Slaves retain their outsider status, and they are excluded from much of the social and political life of those societies.
The reasons for this may be two-fold. As mentioned earlier, Mande societies have long been noted for the existence of endogamous castes-musicians, leather workers and blacksmiths are prominent and recurring examples. They were set apart by occupation and the requirement that they only marry among themselves. Further, they were often characterized by certain traits or activities peculiar to them—e.g., the blacksmiths possessed great spiritual power in smithing, and the musicians were noted for living by begging. Given this form of social organization, it is not difficult to see how slaves could form yet another endogamous clan. In other words, in the logic of Mande social life, slaves could be conceptually accommodated according to a well-established pattern.
The second reason involves Islam in the question of slavery on the Upper Guinea coast. I do not think that this is an easy issue to deal with, but I will offer a tentative analysis of its role.
Allan and Humphrey Fisher wrote an interesting commentary on the travels of Gustav Natchigal. Despite justification of slavery by the classical Muslim texts by Ibn Abi Zayd (10th century) and the Khalil ibn Ishaq (14th century), Nachtigal (and Fisher and Fisher) believe that slavery was as much due to African custom as it was to Muslim law. From the evidence presented in this paper it seems that in Sierra Leone that there is a form of slavery, and an intensity of slaveholding, that varied between Muslim and non-Muslim areas. In what sense, then is Islam related to these forms of slavery?
Islam, as a historic phenomenon, has always known of slavery and approved of it. Arab slavers were the scourge of much of Africa, both east and west, into this century. Furthermore, some of the last instances of slavery were found among Muslim peoples—Mauritania was one of the last places to outlaw it, around 1980.
The more germane question for this essay, however, is how it is, or came to be, that Islam promoted a form of slavery among certain African peoples that is different from that of non-Muslim African societies. Islam, as a world religion ultimately foreign to Africa, makes certain absolute distinctions about the universe and its order. The distinction between the deity and everything else, between the Creator and the created, is certainly fundamental. Similarly, as a religion, Islam distinguishes between believers and unbelievers (and in this second category distinguishes between the “people of the book,” such as Jews and Christians, and pagans). As a social system, Islam distinguishes between men and women, freemen and slaves, those of good birth (such as the family of the Prophet) and others.
Islam on every level, from the theological to the social, is a religious and cultural system which, perhaps because of its semitic origins, categorizes and distinguishes objects of the universe.
Such a system, translated into an African (and, more specifically, a sahelian) context, tended to create—or reinforce or exaggerate—social distinctions: In Mande societies—such as that of the Susu or Mandingo, there was, most probably, a hierarchical social system long before they were islamized. In such a context Islamic influence appears to have accentuated or solidified pre-existing social forms. In other situations, Islam may have strong influence—for example, on the Muslim Vai, because their cultural cousins, the non-Muslim Kono, appear to have neither the number of slaves nor the social customs relating to slaves.
In any case, it appears that Islam provided, however erroneously, for certain peoples, an ideology for the continued segregation of slaves from the normal social life of freemen, even after the abolition of slavery. As with the Susu, such segregation may have nothing to do with an conscious articulation of Islamic ideology, but the del.iberate, and nearly pervasive, distinctions between categories and types of peoples which Islam makes may have perpetrated this sahel ian mode of segregation, even after such peoples moved into the forest region and mingled with the indigenous peoples there.
In conclusion, it seems that the persistence of slavery among the peoples of the Guinea Coast is due to two related factors—pre-existing social organization and islamic ideology. Long after slavery was abolished (and even longer after it had any economic importance), among Muslim Mande peoples a clear pattern of distinction and discrimination between freemen and slaves still exists. Among these peoples there are differences in their treatment of slaves; as shown in this paper, such differences have little to do with factors of economics, location or even Meillassoux's schema of forest and sahelian patterns of slavery. What is crucial, I believe, in understanding the persistence of slave status and role is the nature of Mande caste social organization and its justification (usually unconscious) by islamic ideology.

References
Arkley, Robert, 1965. Slavery in Sierra Leone. Unpublished M.A. thesis, Columbia University.
Coupland, Reginald, 1933. The British Anti-Slavery Movement. London.
Fisher, Allan and Fisher, Humphrey, 1971. Slavery and Muslim Society in Africa. New York.
Fyfe, Christopher, 1964. Sierra Leone Inheritance. Oxford.
Grace, John J. 1977. “Slavery and Emancipation Among the Mende in Sierra Leone.” In Miers & Kopytoff (eds.), Slavery in Africa. Madison, Wisconsin.
Holsoe, Svend, 1977. “Slavery and Economic Response Among the Vai (Liberia and Sierra Leone.” In Miers & Kopytoff (eds.), Slavery in Africa. Madison, Wisconsin.
Klein, Martin A., 1978. Review Article: “A Study of Slavery in Africa.” Journal of African History XIX (4).
Lipschutz, Mark, 1973. Northeast Sierra Leone After 1884: Responses to the Samorian Invasions and British Colonialism. Unpublished Ph.D. dissertation, University of California-Los Angeles.
MacCormack, Carol P., 1977. “Wono: Institutionalized Dependency in Sherbro Descent Groups.” In Miers & Kopytoff (eds.), Slavery in Africa. Madison, Wisconsin.
Meillassoux, Claude (ed.), 1975. L'esclavage en Afrique precoloniale. Paris.
Mouser, Bruce (ed.), 1979. Guinea Journals: Journeys into Guinea-Conakry during the Sierra Leone Phase, 1800-1821. Washington, D.C.
P.P.: Great Britain, Parlimentary Papers, 1899. Report by Her Majesty's Commission and correspondence on the subject of the Insurrection in the Sierra Leone Protectorate.
P.P.: Great Britain, Parlimentary Papers, 1928. Correspondence relating to domestic slavery in the Sierra Leone Protectorate. vol. 18, Cmd. 3020.
Person, Yves, 1971. “Ethnic Movements and Acculturation in Upper Guinea since the Fifteenth Century.” International Journal of African Historical Studies, vol. 4.
Rodney, Walter, 1970. A History of the Upper Guinea Coast, 1545 to 1800. London.
Sanneh, Lamin, 1976. “Slavery, Islam and the Jakhanke People of West Africa.” Africa, vol. 46.
Skinner , David, 1980.Thomas George Lawson: African Historian and Administrator in Sierra Leone. Stanford, California.
Thayer, James S., 1981. Religion and Social Organization Among a West African Muslim People: the Susu of Sierra Leone. Unpublished Ph.D. dissertation, the University of Michigan, Ann Arbor.
Vaughan, James, 1977. “Mafakur: A Limbic Institution of the Margi (Nigeria),” in Miers & Kopytoff (eds.), Slavery in Africa. Madison, Wisconsin.
Willis, J.R., 1979. “Introduction.” In Willis, J.R. (ed.), Studies in West African Islam, vol. 1: The Cultivators of Islam. London.