The production of witchcraft/witchcraft as production: memory, modernity, and the slave trade in Sierra Leone

Rosalind Shaw

Research in Africa has showcased that there is a widespread fear of and preoccupation with witchcraft in both the rural and the urban settings, even despite the presence of modernisation and the changes it has brought to people in regards to their cultural lives, since modernisation does not appear to decrease the cases of witchcraft. In this context, witchcraft and accusations are actually said to flourish, since the social and economic inequalities in rural South Africa provides the setting for witch suspicion and accusation, where the rich accuse the poorer of bewitching them, while the latter accuse the former of having prospered through witchcraft. Furthermore, formal education and labour migration may also contribute to the growth of witchcraft accusations, by exposing people to new ways of thinking and conduct. Here, when migrants return to their homesteads there may be a conflict of values, leading to strains and tensions that eventually result in suspicions and accusations of witchcraft However, the supposed problem of witchcraft has elicited responses at both the community and government levels. In the urban setting people may erect security gates and doors and hire anti-witchcraft experts to protect their property. The accused witches may be expelled from the community, a position often enforced by the threat of violence or destruction of the alleged witch’s property, rather than official positions. Here, the violent community responses have been used to justify government and police intervention. Yet, this has no less reduced people’s worries about witchcraft, since government intervention has created the impression that the authorities are more interested in protecting witches than their victims. Thus, people believe that witchcraft has increased and that government intervention is to blame, and today communities have therefore invented ingenious responses to the continuing threat of witchcraft. During the 1940s and 1950s it was widely assumed that, by forbidding anti-witchcraft action, legislation had led to an increase in witchcraft. From this, it is clear that research needs to be conducted on the presence of witchcraft in the urban setting and how it influences this setting in regards to development and how the community is said to have violent tendencies towards assumed witches in attempts to remove the threat from the society.

Turning my frame of inquiry toward academia, I analyse academic presentations on topics of ‘witchcraft’ conveyed to majority geographer, majority-white and majority-Northern audiences. I argue that (a) the imagined geographies of ‘witchcraft’ have been central to colonial Othering; (b) ‘witchcraft’ is often relegated to the academic periphery; (c) ‘witchcraft’ ontologies are plural, fluid and ambiguous and (d) ‘witchcraft’ is often defiant of academicisations. Given this milieu, scholars risk reproducing a ‘witchcraft’ that is Othering, even when/if the scholarship has an anti-colonial orientation. I call for greater attention to the broadening of academic conference exchanges to allow space for pluriform knowledges

the production of witchcraft/witchcraft as production: memory, modernity, and the slave trade in Sierra Leone ROSALIND SHAW— Tufts University the global witch-city [Hjistorical forms of power became nol merely the means of coercion and subjection but (more interestingly) the conditions for creating particular potentialities—individual, social, and cultural. —Talal Asad, Genealogies of Religion: Discipline and Reasons of Power in Christianity and Islam In this article I address the relationship between memory and representations of modernity in the social imaginary. As an anthropological literature on globalization, modernity, and their representations develops, it is easy to forget that many of the transregional processes associated with modernity are, in perhaps most regions of the world, several centuries old. Representations of late 20th-century global configurations may in fact be shaped by images and metaphors from earlier historical processes, reproducing these as memories that tie together past and present experiences of transformative global flows. One such set of memories may, I suggest, be embedded in discourses of witchcraft and practices of witchfinding in parts of Africa. While pernicious images of African witchcraft have long been integral to Western "inventions of Africa" (Mudimbe 1988), witchcraft is not only thoroughly "modern" (Comaroff and Comaroff 1993), but in some areas may also be, I shall argue, a product of the history of the Atlantic slave trade. We are by now familiar with arguments that "much of what has been taken to be timeless tradition is, in fact, the paradoxical effect of colonial rule" (Dirks 1992a:8)—constructions, for example, of caste in India (Dirks 1992b) and of "Bushmen" in the Kalahari desert (Wilmsen 1989). For globalization, as Gupta and Ferguson observe, is "a shared historical process that differentiates the world as it connects it" (1992:16), generating many of the practices and identities that are taken to be defining emblems of "a culture." In this article, I seek to push back such historical outcomes to a time several centuries before colonial rule by exploring the production of witchcraft in Temne-speaking communities in northern Sierra Leone, tracing connections among witchfinding rituals, the Atlantic trade, and the global circulations that this trade entailed. In discussions of witchcraft (m-ser) in Temne-speaking communities I knew,1 I often heard references to an invisible region of space called the "Place of Witches" (ro-senn). This region, depicted as all around us yet hidden from us, connects Temne towns and villages to a limitless In this article, I examine how memory is entailed in the constitution and representation of global modernity. During four centuries of Atlantic slave trading in Sierra Leone, Temne witchfinding developed as a technique for the ritual production of slaves. Today, Sierra Leone's integration into an earlier Atlantic world is "remembered" both in witchfinding ritual and in cosmological images of an ultramodern witch-city built from the extraction of human life and value. I memory, modernity, witchcraft, slave trade, West Africa, Temne] American Ethnologist 24(4):856-876. Copyright © 1997, American Anthropological Association. 856 american ethnologist urban world of wealth and rapid global mobility. Together with the "Place of the Dead" (ro-karfi) and the "Place of Spirits" (ro-soki), it is one of three invisible regions of space in Temne cosmology that intersect with the visible world of ordinary human beings (m-ru; see Littlejohn 1960b and 1963). No one I asked gave the Place of the Dead or the Place of the Spirits much specific description, but it was a different matter with the Place of Witches: people I asked about witchcraft—whether rural farmers or members of the urban middle class, Muslims or Christians, men or women—usually drew vivid and remarkably consistent images. They often described a prosperous city where skyscrapers adjoin houses of gold and diamonds; where Mercedes-Benzes are driven down fine roads; where street vendors roast "beefsticks" (kebabs) of human meat; where boutiques sell stylish "witch gowns" that transform their wearers into animal predators in the human world (no-ru); where electronics stores sell tape recorders and televisions (and, more recently, VCRs and computers); and where witch airports dispatch witch planes—planes so fast, I was once told, that "they can fly to London and back within an hour"—to destinations all around the globe. What are we to make of these images of affluent modernity, ultrarapid international travel, high-tech media, and costly imported commodities—but also of animal predators and cannibalistic consumption? Their evocation of a modern globalized topography—of time-space compression in "a multidimensional global space with unbounded, often discontinuous and interpenetrating sub-spaces" (Kearney 1995:549)—is startling. Vet this unbounded and alluring global space is not one that ordinary human beings can inhabit. Although its presence may be recognized everywhere, its incompatibility with moral personhood and community is registered both in its intangibility and in its tropes of perverted and predatory consumption: the agency of the witch is necessary for the pursuit of one's desires within it. As in many other postcolonies, such tropes make a powerful popular commentary on a "politics of the belly" through which the state feeds off its own citizens (Bayart 1993; Mbembe 1992)—a commentary that is especially salient in the context of a succession of "kleptocratic" regimes, elite affluence, economic decline, and, since 1991, a devastating rebel war in Sierra Leone (Richards 1996). It is common knowledge that the nation's mineral wealth has hemorrhaged into the overseas bank accounts of politicians and of local and foreign business tycoons, who have indeed built houses in foreign cities out of Sierra Leone's diamonds. And the juxtaposition of images of mobility, luxury, consumption, and predation on the witch-city's streets speaks eloquently of the interconnections between two striking statistics: Sierra Leone has both one of the highest rates of infant mortality and one of the highest percentages of per capita ownership of Mercedes-Benz cars in the world. But these images also, I suggest, have a far deeper ancestry, drawing on representations that became pervasive as part of a much older globalizing process. Specifically, I shall argue that the invisible witch-city may be seen as the "presencing" of a temporally distanced transregional process (cf. Munn 1990) in which persons were turned into trade goods and "consumed" on a massive scale: the Atlantic slave trade. witchcraft, modernity, and history Until recently, little attention has been paid to the historical production of witchcraft in Africa. Whether it was analyzed in terms of "primitive thought" and multiple rationalities (EvansPritchard 1937),2 sanctions against antisocial behavior (Gluckman 1956:92-93), the orderly release of tensions in social relations (Turner 1957), notions of a "limited good" in contexts of economic change (Foster 1965), or ambiguities in local social dynamics (Douglas 1970; Cluckman 1956:97-101), African witchcraft used to betaken for granted as part of a cultural corpus of "traditional beliefs." Recently, however, a different strand of scholarship has located witchcraft and sorcery in contexts with which, it had long been assumed, they could not be coeval: the postcolonial state (e.g., Ciekawy 1992; Geschiere 1988; Geschiere 1994; Rowlands production of witchcraft/witchcraft as production 857 and Warmer 1988); literate, Westernized, urban worlds (e.g., Bastian 1993; Meyer 1995; White 1990); and capitalist enterprise and labor migration (e.g., Apter 1993; Auslander 1993; Matory 1994; Piot 1996b; Weiss 1996). As embodiments of predation and of the "cannibalistic" consumption of human productive and reproductive potential, new forms of witchcraft and vampirism crystallize perceptions of the alienation of labor under colonial and postcolonial capitalism, social consequences of a cash economy, contradictions of urban and international migration, disparities generated by cash cropping, and dislocations of boom-and-bust cycles to which communities and regions are subject through integration into global markets. In this new scholarship, modern forms of witchcraft are usually linked either to postcolonial predicaments or to colonial processes that go back to the first half of the 20th century. The point of releasing witchcraft from its ethnographic atemporality, however, would be lost if the "modern" forms traced in this scholarship were read as standing in opposition to a contrasting category of "traditional" or "precolonial" witchcraft: "probably less is known," as Comaroff and Comaroff comment, "about [witchcraft's] 'traditional' workings than anthropologists often suppose" (1993:xxviii). Austen suggests, specifically, that some of its apparently "traditional" workings in Africa may in fact be products of the slave trade: The conception of witchcraft as an ambiguous attribute of power within Africa is often presented in ahistorical terms, as a timeless reflection of the tension between communal values and selfish individualism and anxieties about natural threats to subsistence. Our data on witch beliefs, however, are all relatively recent; with little exception they are drawn from societies that had long been involved with either the Islamic or European outside world. It is striking that several West Central African cosmologies link witchcraft with the deployment of victims in a nocturnal and/or distant "second universe," echoing, in more or less explicit terms, the experience of the Atlantic slave trade. [Austen 1993:92] Such echoes are especially clear in MacGaffey's work on BaKongo religion and history in lower Zaire (now Congo). MacGaffey tells us, for example, that here "the historical slave trade was assimilated to beliefs in contemporary witchcraft, both regarded as illicit traffic for profit in the souls of persons killed and transported by occult means" (1986:62). Even more intriguingly, he later tells us: Having stolen a soul, a witch may use it to pay his debts to other witches, or he may imprison it in a container. . . and force it to work for its master. . . .A person in this kind of captivity is called the "soldier" (soldat) of his witch master. His soul may be enclosed in a charm so that his anger at the way he has been treated may be turned against his master's victims. Or he may be sold to another master or even shipped to America to be put to work in factories making textiles and automobiles. [1986:162] Given this imagery of the capture, transportation, and forced labor of the soul, we might ask to what extent the slave trade was assimilated to preexisting understandings of witchcraft, and to what extent the historical experience of the slave trade itself produced many of these understandings of witchcraft? Like BaKongo witchcraft cosmology, Temne representations of the invisible witch-city depict a different universe in which plundered value from ordinary human inhabitants of the visible world (no-ru) is deployed by witches. Let us now examine Temne witchcraft in more detail. varieties of Temne witchcraft When talking about witchcraft (ra-ser), people usually describe the hidden acts of a witch (user) whose spirit (an-yina) or shadow (ao-rrunpal) leaves its body during dreams to go on invisible, nocturnal predations (Shaw 1992). Witches typically attack their victims by sucking the blood that holds a person's life (an-nesam) and by removing the heart (ka-buth), the inner center (ao-rjesmsr) of that life and the seat of thoughts, intentions, feelings, and memories (Shaw in press). The blood and heart may be stored in the Place of Witches for later consumption and, until the heart is eaten, the victim will live—albeit as an empty shell. "The baby may appear healthy in the morning," I was told by a young man, "but it will only be its flesh (ao-der) running 858 american ethnologist about." Victims are collectively consumed at meetings of witches' rotating credit associations (e-susufi in the Place of Witches, each member in turn contributing a victim to the group's regular feasts. Members of these "witch credit associations" sometimes recruit others—often children or other kin—by giving them meat in their dreams: parents, accordingly, instruct children to refuse such offerings, explaining that any meat cooked or consumed in dreams is human (Shaw 1992). Witches develop a second stomach Ouj-kuotha) for the digestion of human flesh that an ordinary stomach cannot process: an ordinary human body, by implication, is a moral entity incapable of reducing other people to meat. Witches also have four eyes (e-for t-anle): two ordinary, visible eyes complemented by two invisible eyes that impart a special, penetrative vision (Littlejohn 1960b; 1963). With their four eyes and two stomachs, witches could be said to grow a partial second body—a body of predatory gaze, perverted consumption, and no heart—that removes them from a shared bodily empathy with their victims. The victims of witches' invisible "eating" are usually children (including those still in the womb, not to speak of the womb itself too-pom]—it is the source of future children) or, more rarely, elderly people. Most witches are popularly represented—by women as well as men—as female, and often as wives in polygynous households, who contribute their co-wives' children or their own children for consumption at the feasts of their witch credit association. Healthy adults, who are too strong to be susceptible to invisible vampirism and cannibalism, can be attacked through other types of roser. As we have already seen through the stories of witch gowns bought in the invisible witch-city, witches can also shape-shift (lafthe or folne) into dangerous wild animals of the bush—leopards, snakes, chimpanzees, elephants, and crocodiles—and attack their victims in these forms (cf. Jackson 1989:102-118). They can also beat their victims with sticks and whips that can be felt but not seen, rendering the victims weak or paralyzed. Alternatively, they can invisibly steal growing rice through a practice called ka-ttvfi, which usually involves placing the substance of a crop on an invisible mat and transferring it either from one farm to another or from the human world to the Place of Witches, thereby contributing to the tatter's plenty. This causes poor yields of rice, the crop that is not only the mainstay of a rural household's subsistence but is also symbolically equivalent to people themselves in many contexts (see Littlejohn 1960a; Richards 1986). Through another version of ka-thofi that English-speaking Temne translate as "economic witchcraft," witches invisibly transfer money from other people's pockets to their own or accumulate it in the witch-city. Rumor has it that this is widely practiced by wealthy traders—especially Lebanese traders, who achieved rapid prosperity in both small- and large-scale commerce in Sierra Leone from the end of the 19th century on. It was also described to me as endemic in the diamond-digging district in Kono, in eastern Sierra Leone. The most effective and deadly form of ra-ser, however, is the use of the witch gun (an-pinkar a-serso), a weapon constructed out of such materials as a piece of papaya stalk (or other tube-like object), a grain of sand (or other tiny potential missile), and an explosive powder to activate it. A witch can take these objects and transform them into a piece of lethal artillery that can be used to shoot victims from considerable distances. This kind of witchcraft is depicted as practiced particularly by men who are powerful, affluent "big persons" (an-fem a-bana)—usually men, precisely those depicted as owning the houses of diamonds and the Mercedes-Benzes in the witch-city. These guns, wielded by the most powerful witches, nevertheless form only part of the technology of the Place of Witches: the invisible city is described as the origin of all technological innovations in our own visible world. This claim entails a moral ambivalence about the powers and products of Western technology—an attitude that was conveyed in many of the witchcraft stories circulating during my stays in Sierra Leone.4 In a northern Temne town some years ago, for instance, rumors spread about the disconcerting skills of a Temne-Lebanese teenager. A former schoolmate of his described him in these terms: production of witchcraft/witchcraft as production 859 An Englishman wanted to take him to England but his parents didn't agree. So he read books on his own and made radios and built lanterns.5 He earned money by building radios. He could make a lantern with wonderful electronics—there could be movement inside it. He could make "robot" lanterns. So he has four eyes; he is a witch. He made a wooden airplane and a little engine, and it could fly. His mother was angry and broke it; she was frightened in case he was a witch. The narrator claimed that the boy had "four eyes." All witches, as we have seen, have "four eyes" and, although not all those with this special vision prey on others, they all have the capacity to activate this predatory potential if they wish. For the boy's mother, apparently, her son's identity as a witch was a frightening possibility that might still be arrested by her actions. That she is described as having destroyed a miniature working model of a plane is significant, as witches are said to be able to take miniature objects (usually groundnut shells) and transform them into full-size planes and boats. The fact that the Englishman's offer was rejected is also significant. From a European (and a middle-class Sierra Leonean) perspective, the Englishman would be understood as having wanted to give this brilliant child educational opportunities abroad. But if we bear in mind the image of the invisible witch-city, a more disturbing interpretation is possible: the Englishman's interest might also be seen as a wish to take the boy to a place where he would be able to develop his strange witch capacities further. Like the Place of Witches, European and North American cities represent inaccessible urban landscapes of wealth, power, commodities, technology, mobility—and witchcraft. These links among Europeans, Western technology, and witchcraft are often made explicit in commentaries on imported high-tech consumer goods. Michael, a young man in Freetown, remembered the reaction of his aunt several years earlier, on a visit from her home in a provincial town, to the new television in his family's parlor: My aunt came from Magburaka. She saw our television in Freetown, and asked me if that person was inside the television. I said no; in Freetown there is a big camera with which they are "snapping" the person you see in this television. And this camera sends out the picture. So if you have a television you get the picture and you will be able to see. She said, "Ala! The white men are witches (an-potho a-serf. The white men take their witchcraft out into the world so everybody sees it!" European witches are represented as making their products public and visible, of having harnessed their powers of witchcraft to achieve material success and technological dominance in the world. African witches, in contrast, are represented as keeping their "wonderful" (Temne ninis, a term conveying both marvel and horror) inventions hidden and failing to make them available to their communities, thereby blocking Africa's material development. This connection with economic and technological development is often drawn in discussions of witchcraft, as it was at the end of my 1978 fieldwork, when a man was arrested in Freetown for having hired a diviner to kill his brother with a witch gun. A journalist reporting the arrest in a newspaper article lamented the use of such "scientific knowledge" for purposes of destruction; his final sentence is one that recurred in many discussions of witchcraft I heard: What is this witch-gun? Does it possess a butt, barrell [sic], stock, safety catch and trigger, like any other gun? Exactly what goes to make its bullets? If we could put such scientific knowledge to good use, what a great continent Africa would have been! ["Look Cron" 1978] Such understandings represent "development" as requiring, in a sense, more witchcraft—more accessible witchcraft, benefiting a greater number of people—rather than less. Yet as the account of the inventive teenager and his mother indicates, these yearnings for the benefits of development do not mean that Western science and technology are always viewed with untrammeled admiration. Their categorization as witchcraft indicates, in fact, their ambivalent moral status, as well as that of the Europeans who produce them. European witches may "take their witchcraft out into the world so everybody sees it," but Europeans are not represented as sharing much of it, or indeed much of anything, with others. Indeed, the Temne word for "Europeanness" or "Western ways" (ma-potho) designates a range of antisocial behavior—living a secluded life, exchanging abrupt greetings, not stopping to talk, not visiting 860 american ethnologist others, eating large quantities of meat, and eating alone without inviting others—all of which are defining features of Europeans in Temne experience (Turay 1971:338). Such qualities of seclusion, selfishness, and greed are also precisely those that characterize witches. In this sense, then, witchcraft may be seen as a form of "failed science," although not in the sense imagined by 19th- and 20th-century European intellectualists writing about "primitive thought": the failure here is one of morality, not of rationality. Like BaKongo representations of stolen souls laboring for witch masters in the American car and textile factories to which they have been transported, the gleaming but troubling images of the invisible witch-city provide a commentary on problematic forms of wealth, power, technology, and mobility. Both sets of images evoke a vision of "a totalizing moral economy . . . in which the acquisition of money and power links parochial means for producing and destroying human value to potent foreign sources of wealth" (Comaroff and Comaroff 1993:xxv). Let us turn, finally, to the historical beginnings of this link between potent foreign wealth and the consumption of human productive value in this part of Africa. the production of witchcraft While late 20th-century Temne representations of witchcraft link the "cannibalistic" consumption of persons to wealth, imports, and technological power, for nearly four of the five preceding centuries people had, in fact, been reduced to consumable entities and exchanged for foreign wealth in the Atlantic slave trade. The written sources for these centuries do not tell us much about representations of witchcraft, but they do describe in some detail the practices of public divination by which persons were transformed into slaves through their conviction as witches. When the Portuguese and other European traders arrived on the part of the Upper Guinea Coast they named "Serra Lyoa" ("Lion Uplands") in the latter half of the 15th century,6 they began to exchange goods such as iron bars, woven textiles, and salt for ivory, gold, kola, and—most lucrative of all—slaves. Translocal trade was not new in this region, however. Well before the arrival of Europeans, the peoples of the Upper Guinea Coast had been organized into commercial networks that connected their coast and rivers together and linked them to long-distance trade routes leading to the Mande kingdoms of ancient Ghana and Mali to the northeast (Brooks 1993:79-96). The coming of European trade, then, did not suddenly propel a formerly isolated and self-contained area into a wider sphere of commercial (and hierarchical) relations (see Gupta and Ferguson 1992); rather, a new kind of transregional circulation and exchange became integrated with already established translocal forms. Our earliest surviving, detailed account of Sierra Leone comes from these late 15th-century beginnings of the trade with Europeans. It was written by Valentim Fernandes, who is thought to have obtained his information primarily from Alvaro Velho, a Portuguese who lived on or near the coast of Sierra Leone for eight years, probably as a trader (Fernandes 1951 [1506-10]:103; 175-176, n.197). Fernandes (or Velho) tells us that this part of the coast was inhabited by Bullom-speakers (Boldes) and, further inland, by Temne-speakers (Temjnis).7 Both peoples lived in villages, each village having a king ( See in Bullom; Obe in Temne) who was paid no tribute, but who was owed farm and building labor, assistance in war (provided that a council of elders agreed the war was just), and special parts of certain wild animals killed in his domain (Fernandes 1951 [1506-10]:80-83).8 In his fairly rich account of shrines, images ("idols"), spirits, ancestors, sacrifices, burial practices, divination, and healing (Fernandes 1951 [1506-101:80-93; Fyfe 1964:27-28), Fernandes makes no mention of witchcraft or techniques of witchfinding. His descriptions of divination for sickness, remedial sacrifice, and burial practices are worth quoting for the contrast they provide to later accounts I shall examine below: production of witchcraft/witchcraft as production 861 In the case of sickness they have no medicine nor know its rules, and only if someone is sick they go to the fetish-men (feyticeyros) who throw lots to know why he is sick; they always find that it is the idols (os ydolos) who send the sickness, and order that to him should be sacrificed a dog, a goat, or a chicken, for these are the animals that they are accustomed to sacrifice. . . . Here is how they conduct their burials, and when a noble dies, or an honored man, they open his side and take out all the entrails; they wash them very well, and fill him with herbs like mint that smell very nice, and throw inside him rice-flour and rub him with palm oil, and make him a platform of 5 or 6 steps high, decorated with the best cloths there are in the area, like red or cotton cloths. Then they seat the dead man in a chair wearing the best clothes he has. .. . The day of the burial all the people gather in the place where the dead man is on the platform, and silence is ordered. And all give jewels according to their means to whoever they wish, with a declaration; if it is to the dead man they are buried with him, if it is to the dead man's relations each takes away what he has been given. (Fernandes 1951 (1506-10:86-91] Additionally, if the deceased was a man who had killed many people in battle, the mourners would "play the drums as if for war" and "parade as they would do if goingoff to war" (Femandes 1951 [1506-10] :90-91). Fernandes also tells us that "the justice of the Teminis is that according to custom thus murder and all other wrongs and offences are compensated for by money and no one is killed" (1951 [1506-10]:96-97). It would be wrong, of course, to conclude from this that witchcraft, witchfinding, and the execution of wrongdoers did not exist in Temne-speaking communities at the turn of the 16th century. Yet it is interesting that, in contrast to their absence from Fernandes's sketch of these early days of the slave trade, later writers represent the ritual identification of witches, followed by their sale or execution, as a routine and highly visible part of practices of divination, healing, and burial. As some historians have noted, several of these later writers specifically attest to a close relationship between witchfinding and the slave trade (Fyfe 1962:9, 1964:75-76; Rodney 1970:107). This connection between the Atlantic trade and witchcraft does not, however, fit i nto a simple, unidirectional history of European "causes" and African "consequences." Fifty years after Fernandes wrote, in the second half of the 16th century, European traders and sailors described what they saw as an "invasion" (a process that historians have interpreted as a more gradual imposition of domination) by a Mande-speaking group called the "Mane" (Jones 1981). This group founded four states along the coast, imposing their form of kingship on local peoples such as Temne-, Bullom-, and Loko-speakers (Brooks 1993:303-304; Rodney 1970:47). Initially, the Mane sold captives to European slavers as a by-product of their warfare (Brooks 1993:293-294). By the turn of the 17th century, however, the supply of slaves to Europeans became an end in itself, and this, according to Rodney (1970:102), gave rise to an escalation in warfare and in the flow of captives sold as slaves: the Mane had become middlemen in the Atlantic trade. Warfare was nevertheless not the only source of the slaves the Mane sold. It was at this time that a Portuguese Jesuit priest, Manuel Alvares, spent ten years in an ultimately unsuccessful mission to Sierra Leone. Alvares wrote a richly detailed account of the peoples—especially Temne-speakers—in the region (1990[c.1615]), in which he describes dominant etiologies of illness and death as attributing these misfortunes to a process of invisible consumption by enemies: "They entertain a malevolent and false belief about their illnesses and physical ailments, for they say, and they sincerely believe this, that God does not send death to them, rather it is their enemies and rivals who are eating them up" 0990[c.1615], pt.2, ch.3:4). Diviners ("sorcerers") gave similar readings of sickness and death9—readings that differed from those of "the fetish-men who throw lots" a century earlier, in Alvaro Velho's time. Instead of "always find[ing] that it is the idols who send the sickness," malevolent human agency was now sought among those close to the patient: They have various other forms of divination, but we need not discuss them all... . When they take to their beds, the first thing they do is to consult Accaron. Like Ocosias, they send messengers to the most famous sorcerers, who put a thousand lies into their heads. The sorcerer, this servant of hell, puts on the fire a small pan containing a little water. . . . (l]n order to find out if anyone is giving him poison or is otherwise responsible for his death, the sorcerer names the man's relatives. If the water sinks back when one is named, he is not to blame; but if it boils over, they consider that the person named is the guilty 862 american ethnologist o n e . . . . If it does this, they fall on the poor man, and on the orders of the king kill him or enslave him. [Alvares 1990(c.1615), pt.2, ch.3:4] No longer solely a process of ritual mediation between "the idols" and the human community amidst which they dwelled, divination and the understandings of affliction it enacted and reproduced were now—under the auspices of the new Mane rulers—integrated into the European slave trade. This integration of divination into the Atlantic system receives more explicit comment in Alvares's account of other coastal peoples further to the north—the "Banhus," the "Biafars," and the "Sonequei"—in what is now Guinea-Bissau and Senegal. In detailing the injustices of slaving, " 'the most substantial trade' of Guinea," he complains that "the heathen . . . enslave large numbers on the grounds that they are witches, an offence which they can only prove by diabolic arts" (1990[c.l61 5], pt. 1, ch. 5:3), such as the ordeals of "red water" poison and of red-hot iron (199O[c.1615j, pt. 1, ch. 7:8-9). Among the Banhus, the placing of a corpse on a platform—a mobile version of that described in Sierra Leone by Fernandes a century earlier as a means of honoring the deceased—was now part of the material apparatus of witch detection and enslavement: When a sick man dies, they straight away ask him not to flee away but to reveal who ate him. . . . Four of the heathen then carry the dead man on a wooden grating like a bier, and sometimes they add the cloth the man wore when he died. They quickly make their way around the village, from one side to the other and through the open places. And whenever the ministers of the devil stop, it is said and falsely believed that the people in that spot "ate" the dead man. . . . The wretch whom the bier accused (by stopping) has to pay. His house is attacked and a host of his children are enslaved. [Alvares 1990(c.1615), pt. 1,ch. 7:13]'° The witchfinding divinations Alvares describes both entail and reproduce particular constructions of social relationships. If someone falls sick, on the one hand, "the sorcerer names the man's relatives" (1990[c.1615], pt. 2. ch. 3:4) and looks for witchcraft among the victim's kin; if the victim dies, on the other hand, and the ritually interrogated corpse stops in front of another person, "a host of his children are enslaved" (1990[c.1615], pt. 2. ch. 3:4) The ties that connect victims to witches, then, as well as witches to other witches, are those of kinship: witches both consume their kin and reproduce themselves through their kin. Alvares also tells us that when those accused belonged to families with means, they stood a much better chance of being exonerated in an ordeal. At that time, he writes, the most common trial by divination was the poison ordeal known as the "red water": They boil the bark of macone, then seat the accused in a public place and give him the poisoned drink, which kills only those who have no one to help them. As I have stated elsewhere, immediately after the person has drunk a calabash of red-water they give him a calabash of plain water to drink. If the accused expels the macone in one vomit, they consider him innocent. But if otherwise, he dies lamentably. The relatives of those who have to submit to the ordeal help them to prepare for it. Most of the accused vomit the poison before it can do them any harm, as a consequence of their taking antidotes beforehand, antidotes such as gold dust, the most commonly employed, or oil. They consult beforehand a sorcerer, to find out the right moment to take the poison, and sometimes they try it out on slaves or hens. So those who have the means thus escape the effects of the ordeal. [1990(c.1615), pt. 2, ch. 3:5) Distinctions of power between those with and without "means," then, were translated into the distinction between those convicted and those absolved, which also determined whether the kin of the accused would also be sold into slavery. This intersection of inter- and intralineage politics (in the accusation of the victim's kin and in the disparities among accusations of members of elite and nonelite lineages) with the economics of the Atlantic trade (in which the sale of whole families generated more wealth than the sale of single individuals) was to become even clearer as the slave trade gathered further momentum. In the 18th century, slaving reached its peak. By this time, English merchants had replaced Portuguese traders in their domination of the slave trade, and the Mane rulers had long been absorbed through intermarriage with their Temne-speaking subjects. These Temne speakers had production of witchcraft/witchcraft as production 863 meanwhile expanded and absorbed other groups, and were now organized into several centralized chiefdoms; like their former Mane overlords, moreover, they had become intermediaries in the slave trade (see Sayers 1927:14). During this century, however, certain external groups—Susu, Fula, Mende, and agents of European slave traders—encroached on (and, in some cases, gained control of) many areas settled by Temne-speaking communities in order to exploit the trade themselves (Ijagbemi 1968:24-75). In addition to the wars that resulted from such external encroachment, wars fought between Temne-speaking communities for internal political reasons were encouraged by agents of European slave-traders, who exacerbated the conflicts by supplying both sides with arms (Ijagbemi 1968:26; Rodney 1970). Captives seized in either kind of war fed the Atlantic market, and in this century warfare and raiding reached unprecedented levels (Rodney 1970:250-51): "[a]s the Atlantic trade expanded," according to Ijagbemi, "so did the slave-motive of the wars mount, and so did the wars increase both in frequency and magnitude" (1968:74). As in the era of Mane "invasion" and rule, the main sources of slaves were warfare and raiding, the divinatory conviction of witches and other wrongdoers, the pawning of dependents, and the seizure of debtors." In the late 18th century John Matthews, an agent for an English slave-trading firm, wrote: The best information I have been able to collect is, that great numbers [of slaves) are prisoners taken in war, and are brought down, fifty or a hundred together, by the black slave merchants; that many are sold for witchcraft, and other real, or imputed, crimes; and are purchased in the country with European goods and salt. 11961(1 7881:146] While Matthews was aware of a connection between witchcraft and the slave trade, however, he sought to justify the latter by arguing that it was both the strength of witchcraft beliefs and what he described as an "addiction" to witchfinding techniques in Sierra Leone that drove the sale of slaves—thereby reversing the main direction of their interrelationship. He wrote: Though most unenlightened nations believe in charms and witchcraft, yet the inhabitants of this country are so much addicted to it that they imagine every thing is under its influence, and every occurrence of life they attribute to that cause. . . . If an alligator destroys any body when washing or swimming, or a leopard commits depredations on their flocks or poultry; if any person is taken suddenly ill, or dies suddenly, or is seized with any disaster they are not accustomed to, it is immediately attributed to witchcraft: and it rarely happens that some person or other is not pointed out by their conjurors, whom they consult on these occasions, as the witch and sold. [1961 (1788):130-132] Despite Matthews's self-serving misattribution of witchfinding to a Sierra Leonean "addiction," his description of practices of witch detection are of considerable interest. The "conjurors" he mentions pointed out "some person or other" through ritual interrogations of the corpse similar to those Alvares had written about, in a different language, nearly 200 years earlier. Six bearers, according to Matthews, carried a bier on which was placed either the deceased or the latter's clothes; the ritual power of the corpse or the clothes controlled the bearers' movements. At the gravesite, a friend or kinsperson stood before the bier, held up a reed, and addressed the funeral litter; "[a|nd then he asks him what made him die—whether he knew of his own death, or whether it was caused by witchcraft or poison"—to which the deceased answered either affirmatively by propelling the bearers forward involuntarily, or negatively through "a rolling motion" (Matthews 1961 (1788]:122-123). Such burial divinations were "constantly practised" {Matthews 1961 (1788]:129). Afterward, distinctions of position and privilege determined the scope of the consequences for the kin of the accused: "the culprit is then seized, and if a witch sold without further ceremony: and it frequently happens if the deceased were a great man, and the accused poor, not only he himself but his whole family are sold together" (Matthews 1961(1788):124). Those accused of poisoning, however, were usually permitted the chance—apparently slim, in contrast to what obtained in Alvares's time—to exonerate themselves by taking the "red water" poison ordeal. The accused would be confined "in such a manner as he can release himself" (Matthews 196111788]:125) and would then "escape" to the next town, declaring to its headman that he wished to prove his innocence by drinking the 864 american ethnologist red water. If the poison killed him he was declared guilty; if he vomited it up he was provisionally declared innocent—but, if he then defecated within a day he was described as having "spoiled the red water" and was accordingly sold (1961:126). It was not only those accused by corpses who had to undergo this ordeal: Matthews was told that in the interior, "suffering the people to drink red water upon every trifling occasion, was attended with such fatal consequences as would in time depopulate the country" (1961 [1788]:129). Writing 15 years later, at the beginning of the 19th century, the traveler Thomas Winterbottom similarly states that explanations in terms of witchcraft were almost universally given for death, and describes the divinatory interrogation of the corpse as a standard part of burial practices (Winterbottom 1969(18031:260, 236). The social disparities observed by Matthews are also noted by Winterbottom as producing divergent etiologies of witchcraft according to the power and status of the deceased. There were two possibi lities: the deceased either identified the witch responsible, or acknowledged having been a witch who was finally killed by the retaliatory agency of the personal spirit (u-/»rft)orcountermedicine of an intended victim. "It is most usual," Winterbottom writes, "to assign the former cause for the sickness and death of chiefs, and other people of consequence and their connections; and the latter for those of any of the lower class" (1969[1803]:236). When the first explanation was revealed by the movements of the corpse's bearers, the deceased identified the witch or witches responsible by selecting from among a series of names. The fate of those accused then rested upon their connections to power: "these," Winterbottom continues, "if they have friends to plead for them, are allowed the privilege of appealing to the red water ordeal in proof of their innocence; but if not, they are sold" (1969[1803]:238). The importance of having a "big person" (w-uni bana) "behind" you as a patron is underscored by these disparate outcomes. Such patrons would usually have been members of the ruling lineages of a town or chiefdom—those who traced descent from ancestors who had cleared the bush to form the original settlement—and who could provide their supporters with protection from warfare and raiding in exchange for labor (Bledsoe 1980). During the slave trade, when every death or serious illness seems to have entailed divinatory accusation as part of routine procedures of healing and burial, such important attachments were likely to have been essential in providing people with "friends to plead for them." Four years after the publication of Winterbottom's account, the Anti-Slave-Trade Act of 1807 was passed by the British Parliament. This did not mean the end of the Atlantic slave trade, since other European nations continued it, but the Act represented the beginning of a process of eradication over the next several decades (Fyfe 1964:155-165). It also represented the gradual decoupling of witchcraft and public divination from the Atlantic system. One contemporary observer, the English traveler Laing, commented on what he saw as a consequence of this disengagement in his journey through a part of Temneland: While I was at Ma Bung, a young girl died rather suddenly; and previous to her interment the following practices were observed. . . . A few hours after the death of the girl, (he Elders and the Greegree men of the town assembled in the palaver-hall, and held a long consultation or inquest as to the probable cause of the death. It was inquired whether any one had threatened her during her life-time, and it was long surmised that she might have been killed by witchcraft. Had the slave trade existed, some unfortunate individual might have been accused and sold into captivity; but its suppression in this country from vicinity to Sierra Leone, permitted the Magi, after a tedious consultation of three days, !o decide that the death had been caused by the agency of the Devil. [1825:85-86] Laing's account is suggestive of a transition in the political economy of knowledge. It suggests that toward the end of the Atlantic trade, as the divinatory accusation of witches ceased to be a means of transforming people into trade goods, etiologies of spirit-attack ("the agency of the Devil") acquired increased salience. These etiologies were not unlike those described by Fernandes when he declared that "it [was] the idols who |sent] the sickness" (Fernandes 1951(1506-10j:89) in the early days of the slave trade more than 300 years before. production of witchcraft/witchcraft as production 865 witchfinding and the state Although length constraints do not allow me to present a detailed account of witchfinding techniques after the slave trade (see Shaw n.d.), a brief outline of that phase of their development may be useful nonetheless. When, at the end of the 19th century, the British colonial authorities imposed a Protectorate over the Sierra Leone hinterland, they also imposed different techniques of truth through the legal apparatus of the colonial state. Rejecting the divinatory practices through which individuals were identified as witches (Public Records Office 1899:142), they nevertheless accepted local definitions of witchcraft as a crime, and proceeded to try witchcraft cases in their District Commissioners' Courts (Fyle 1988:95; Public Records Office 1892-1899). In these colonial witch trials, however, witchcraft was redefined in accordance with rationalist British conceptions of "proof": instead of the invisible "eating" of victims that had been so prominent in witchcraft convictions during the slave trade, emphasis was now placed on "physical evidence" in the form of the accused's possession of harmful ritual substances (see Fyle 1988:95-96). Making this the defining feature of "the truth" meant that the range of activities that constituted "witchcraft" in the District Commissioner's Court was reduced to one side of an axis corresponding to the Western dichotomy between "matter" and "spirit," or between "fact" and "imagination." Ironically, one consequence of these witch trials in colonial courts was to authorize the legal apparatus of the state—the court system, the police, procedures of arrest, fines, and imprisonment—as a proper arena for witchcraft cases. It has retained that authority in postcolonial Sierra Leone, although now such cases are heard in Paramount Chiefs' Courts, which has resulted in the return of control over the conviction of witches to local rulers, from whom it had, at least officially, been removed in the colonial era. Moreover, the colonial emphasis on the "physical evidence" of witchcraft has now yielded to Temne understandings of witchcraft as invisible consumption, and the public divination techniques that alone have the capacity to discern the hidden "truth" of such consumption have accordingly been integrated into these courts.'2 All diviners, especially those who perform public divinations, are now required to get a license from the police.13 Thus authorized, these licensed diviners work with the police, and their oracular findings are presented as "evidence" in court.14 Witchfinding divination has ceased to be a requisite part of healing and burial in contemporary Temne ritual practice, although it is performed when there is a series of deaths (or when single deaths are construed as unusual) in a household or village (see, e.g., Shaw 1985, 1991, and n.d.). Only household heads may arrange such a divination, and only after seeking permission from—and paying a fee to—the Paramount Chief. Since few can afford these expenses (in addition to the Chief's fee, diviners often charge the equivalent of $20 or more for a public divination), this means that chiefs and relatively affluent senior men still monopolize these techniques of truth. Today, however, conviction as a witch no longer leads to social death as a slave in the Atlantic system. Instead, it usually results in a beating, a fine levied in the Paramount Chief's Court, and the application of medicine to the eyes to extinguish the witch's supernaturally doubled vision. Instead of being integrated, via localelites, intothe Atlantic slave trade, public divination is now integrated—again via local elites—into the state. witchcraft as production What potentialities were produced by the imperative to find witches during the slave trade? Many of the subjectivities subsumed by Temne understandings of ra-ser—readings of illness and death as products of invisible consumption, dreams of eating meat, and dispositions toward particular suspicions and accusations—may, I suggest, be seen as such productions. This does not necessarily mean that understandings of witchcraft were absent before the slave trade, but 866 american ethnologist it shows that, as techniques of witch detection became part of routine healing and funeral practices through the impetus of the slave trade, these understandings were driven into elaborate and ubiquitous forms. Temne knowledge of witchcraft, then, was at least partly a product of witnessing witchfinding divinations as they unfolded again and again. Repeated experiences, for example, of a funeral litter carried by bearers, of a man standing in front of a grave, holding up a reed, of questions he addresses to the deceased ("And then he asks him what made him die—whether he knew of his own death, or whether it was caused by witchcraft or poison" [Matthews 1961 (1 788):122-123]), of the bearers lurching forward as the deceased answers "yes," of the rolling motion of the funeral litter to indicate "no" (Matthews 1961 [17881:122-123): such procedures are "read with the whole body" (Bourdieu 1990:76), assimilatingunderstandings of witchcraft as an inevitable and ubiquitous part of lived experience. Included within these understandings are implicit theories of social relationships: ideas of where loyalty and treachery lie, of the location of trust and mistrust, and of axes of conflict and solidarity. As we have seen, accusations and convictions of witchcraft were advanced in particular directions—inward toward kin and downward away from members of ruling lineages and other "big persons." In the literature on witchcraft in Africa, considerable attention has been paid to how witchcraft accusations are generated by tensions within structures of kinship, patterns of authority, and forms of hierarchy (e.g., Evans-Pritchard 1937; Ceschiere 1994; Gluckman 1956; Turner 1957); this is in fact one of the sources of the supposedly "traditional" character of African witchcraft. Yet what if convictions of witchcraft—driven by such transregional historical processes as the Atlantic trade—had a more recursive relationship with these fault-lines of conflict, being not merely their product but also the means of their exacerbation and intensification?15 It was not only kinship but also growing distinctions of class that constituted the relations of power permeating witchcraft accusations. In the early 17th century, it is worth remembering, Alvares tells us that "they boil the bark of macone, then seat the accused in a public place and give him the poisoned drink, which kills only those who have no-one to help them . . . [while] those who have the means . . . escape the effects of the ordeal" (1990[c.1615], pt. 2, ch. 3:5). Being "without means" meant being without either material resources or important links of kinship and marriage—especially connections, we may suppose, to the ruling Mane elite at that time. Nearly 200 years later, in the late 18th century, contrasting trajectories of witchcraft conviction for the well-connected and for those lacking such resources, respectively, were also manifest in terms of distinct etiologies of witchcraft. Recall that, according to Winterbottom, "for the sickness and death of chiefs, and other people of consequence and their connections," the divinatory interrogation of the corpse usually produced a particular person as the witch responsible for the deceased's death (1969(1803]:236). But "for those of any of the lower class" the deceased usually acknowledged, through the same divinatory interrogation, having been a witch who was killed by a counterattack of the personal spirit or medicine protecting a "big person" the deceased had tried to harm (196911803]:236). Dead witches could not be sold, but their living kin could be and often were.16 This last point, however, raises a question about the popular images of the invisible witch-city with which I began. If understandings of witchcraft are produced and reproduced through practices of witch detection in which, under the direction of rulers and members of elite lineages, those accused and convicted have usually been those with less power and wealth, why are "big persons" (an-fam a-bana), with their luxurious properties, sleek vehicles, and high-tech consumer goods, so prominent in representations of the Place of Witches today? Although "big men" were—and still are—rarely formally convicted in witchfinding divinations, they are often rumored to be sinister, witch-gun-toting figures whose power derives from the absorption of medicines made through ritual murder (Ferme 1992; MacCormack 1983; Richards 1995; Shaw 1996)—a form of ritual "cannibalism" firmly within the compass of the Temne category of production of witchcraft/witchcraft as production 867 ra-ser.17 And even though most convicted witches, I was sometimes told, may appear weak and unimportant in this world (no-ru), they are "big persons" in the invisible city. Although powerful, wealthy persons, then, directed witchcraft convictions toward those most unlike them, it is upon them, together with the European world they mediate, that images of the true and hidden form of the witch are modeled.18 For, in addition to driving the production of witches and understandings of witchcraft in particular directions through witchfinding divinations, the Atlantic slave trade itself provided apt metaphors of wrongful "eating" that were deployed in rumors about those who were its most visible agents. Phantasmagoric images of Europeans and of African elites as consumers of human life were reported throughout the centuries of the slave trade. While the 16th-century Mane "invaders" imposed their rule on local peoples and profited from the saleof the latter to European slave-traders, they were themselves the subjects of rumors of grotesque forms of cannibalism (e.g., Alvares 1990[c.1615], pt. 2, ch. 11:1-2; Donelha 1977[1625]:101; Rodney 1970:56). And when, two centuries later, English merchants replaced Portuguese traders in theirdomination of the slave trade, they too—appropriately enough—became the objects of widespread cannibalism rumors that survive to this day (e.g., Matthews 1961 [1788]:152; Joseph Opala, personal communication, October 1989; Rankin 1836:104; South Carolina ETV 1992). Inverting the relationship between accusers and accused described in previous sections, these rumors provide an apt idiom for Mane and European "consumption" of human "commodities" (Shaw 1996; cf. Miller 1976 and 1988). The development of witchfinding as the ritual production of slaves for the Atlantic trade and as the source of particular kinds of knowledge and subjectivity about witchcraft is, then, only part of this story of witchcraft and the slave trade. LikeTaussig's powerful account of Putumayo torture stories as a "colonial mirror that reflects back onto the colonists the barbarity of their own social relations" (1992:164), the very imagery of witchcraft as the wrongful "eating" of people provided counterrepresentations through which those who profited from the ritual conviction of witches were implicitly recognized as the biggest witches of all. 19 Although power may produce knowledge, it does not necessarily determine consciousness. making history: memory and modernity I have argued that in the visions of global modernity materialized by the witch-city, a political economy is "remembered" that once linked divinatory ritual, social relationships, and cosmological representations in Sierra Leonean villages to a commercial system spanning three continents. During the slave-trade centuries, witchcraft and divination in much of the Upper Guinea Coast were part of an expanding Atlantic world in which inflows of European wealth and commodities were tied to the largest process of forced migration in world history. Witchfinding divinations developed as constitutive and persuasive practices that created a reality, like other rituals in other contexts, by "actively refigur[ing| meanings in line with changing perceptions of the universe" (Comaroff and Comaroff 1993:xxi). These forms of divination, then, were more than "ways of knowing" the world (Peek 1991); they actively contributed to its constitution. By authorizing one of the principal ways in which persons were transformed into slaves, those who controlled witchfinding divinations accused those convicted as witches of an invisible "eating" of others analogous to that in which they were themselves engaged. Yet at the same time, their own participation in an "eating" in which the appearance and accumulation of new wealth was engendered by the disappearance of people into oblivion was itself the object of commentary in the form of popular images, stories, and rumors that reversed the direction of formal divinatory accusation. I hope to have shown that we, in turn, may "divine" memories of 868 american ethnologist these processes in representations of the invisible witch-city and in the long-standing rumors of European and elite African cannibalism that they reiterate. To what extent may we also see memories of the slave trade as made present in representations of witchcraft and practices of witchfinding elsewhere in Africa? lam not, of course, suggesting that the Atlantic slave trade took the same form or had the same consequences in other parts of Africa (see, e.g., note 8 below), or that the many disparate understandings of witchcraft in Africa are the outcome of the same historical process. Nevertheless, it would be instructive to examine whether the argument traced in this article has a broader relevance in other regions—perhaps in Congo, where witchcraft and the slave trade are "both regarded as illicit traffic for profit in the souls of persons killed and transported by occult means" (MacGaffey 1986:62), or Angola, where tropes of conquest and enslavement as "eating" were especially prominent (Miller 1976, 1988). A serious exploration of this question would take me beyond both the scope of this article and the limits of my knowledge, but I hope others will address it. The time depth encompassed by such speculations demonstrates the limitations of a concentration on global flows as largely recent "transnational" phenomena, born of nation-states and their borders, of multinational corporations and electronic media. It also demonstrates the necessity of moving beyond the dualistic temporal categories of "precolonial" versus "colonial" (or "postcolonial") in anthropological discussions of modernity, globalization, andtransnationalism (see Cable 1995)20—especially for West Africa, in most parts of which the colonial era did not stretch back much before the late 19th century.21 Focusing on colonialism alone and consigning what came before it to a homogenized "precolonial" past replicates earlier, ahistorical studies by pushing the temporal blind spot back 100 years or so. The idea that it is through particular histories that "local" visions and appropriations of modernity are constituted is not new (e.g., Sahlins 1988). Usually, however, such ideas entail the assumption that the main source of plural modernities is a cultural specificity derived from autonomous ("precolonial") historical trajectories—an assumption that Daniel Miller (1995) criticizes for its valorizing of a priori assumptions of cultural diversity and its disregard of novel "a posteriori" differences in the domestication of global forms. Yet, as the relationship between witchcraft and the slave trade in Sierra Leone demonstrates, in some parts of the world the transformation into a universe of foreign commodities, class, and global commercial flows dates back to the late 15th century. To emphasize memory in modernity is neither to give primacy to the "premodern" nor to dismiss the agency and creativity behind new "local" forms of "global" institutions, but simply serves to remind us of the importance of reproduction, structure, and historical consciousness in all agency (e.g., Bourdieu 1990:54-58; Giddens 1979:49-95, 1981:34-37). What, finally, can we learn about globalization and modernity from the witch-city? One reminder that this ubiquitous but invisible city gives us is that an unbounded global space exists mainly as a function of specific kinds of power and capacity; that "there are continuous paths that lead from the local to the global, from the circumstantial to the universal,... only so long as the branch lines are paid for" (Latour 1993:117). We learn that this global city is in many ways an eminently desirable space for those whose "branch lines are paid for": they need not hunger for meat or money; they have special capacities for rapid mobility; the most irresistible commodities may be theirs. But we also receive clear instruction as to what constitutes payment for this privileged access—the lives and wealth of others, misappropriated from an everyday African world. For all its allure, the witch-city counters the poetics of optimism that may be sensed in much anthropological writing on transnationalism and globalism by a "poetics of predation" (Comaroff and Comaroff 1993:xxvi) that equates global flows with the flow of blood. Built in the Temne imagination through an extended historical experience that tied transregional commercial flows and foreign commodities to a traffic in human lives, the witch-city is a Sierra production of witchcraft/witchcraft as production 869 Leonean "cosmology of capitalism" (Sahlins 1988) that "remembers" this experience in its construction as a startling commentary on global modernity. notes [Editor's Note: The Editor wishes to note for the record that this article was submitted before the author was appointed to the position of Book Reviews Editor for AE. Although the article was accepted after Professor Shaw's appointment, this publication decision was at the unanimous recommendation of four external, anonymous reviewers.! Acknowledgments. My biggest thanks go to those who gave me their hospitality and assistance in Sierra Leone. My field research in Sierra Leone was funded by the former Social Science Research Council of the U.K. (1977-78), the Royal Anthropological Institute's Horniman Scholarship (1977-78), the Carnegie Trust (1989), the University of Edinburgh's Centre for African Studies (1989), the Wenner-Cren Foundation (1992), and Tufts University (1992). I am grateful for discussions of earlier versions of this article at the 1995 Satterthwaite Colloquium on African Religion and Ritual, the 1996 Annual Spring Meeting of the American Ethnological Society, and the graduate African History seminar at the University of Texas at Austin in January 1997. Special thanks are due to Jeanne Penvenne and David Beach for their translation of Portuguese excerpts from Valentim Fernandes's Description of the West Coast of Africa. I am also indebted to Ralph Austen, Misty Bastian, John Comaroff, Toyin Falola, Christopher Fyfe, Christine Ward Cailey, Cory Kratz, Michael Lambek, Michael McCovern, Adeline Masquelier, and Stephan Palmie for their insightful readings. Finally, the anonymous reviewers for American Ethnologist gave me particularly important and valuable comments, for which I am extremely grateful. 1. My fieldwork was carried out in the towns of Matotoka (1977), Magburaka (1977), Makeni (1978), and Petbana(1978,1989) in the Northern Province; in Kono in the Eastern Province (1978); and in Freetown (1978, 1989, and 1992). 2. That it was the rationality of Azande witchcraft (as opposed to any other Zande cultural ideas and practices) that Evans-Pritchard used in order to refute Levy-Bruhl's ideas about "primitive mentality" is, of course, very telling in view of the fact that few cultural forms have been so strongly charged as witchcraft with Western meanings of "the irrational" and "the primitive." 3. Rotating credit associations (e-susu) are mutual aid groups formed to help members save money on a regular basis. For each meeting, members are required to bring a certain sum of money and, once these contributions have been allowed to accumulate, a lump sum is given to one member. This process of contribution and accumulation is repeated so that each member in turn builds up an investment for either long-term goals (such as starting a business) or emergencies (such as funeral expenses). In witches' e-susu, each member in turn is required to bring a human victim to the meeting; if no victim is contributed, the group falls on and consumes the delinquent member instead. Thus, it is specifically the kind of association that supports people in acquiring habits of capitalist accumulation that provides the ties that connect witches together. 4. For similar connections between European technologies and witchcraft (or other sinister occult forces) in other parts of Africa, see Auslander 1993; Bastian 1990; Kaspin 1993; Masquelier 1992; Meyer 1995; Weiss 1996:202-219; White 1990. 5. "Lanterns" are mobile art forms carried in a parade in towns and cities during an annual festival called on-wuthpa, marking the end of Ramadan. 6. From the late 15th century to 1808, "Sierra Leone" refers to the part of the Upper Guinea Coast that the Portuguese named "Serra Lyoa." From 1808 it refers to the British Crown Colony established on that coast, based in Freetown; and from 1896, it was also used of the British Protectorate of Sierra Leone, which was imposed on what was previously termed the "hinterland." From 1961 it refers to the nation-state of Sierra Leone. 7. As with all the peoples of this region, the name "Temne" represented a language group rather than a fixed ethnic identity. It was not until the late 19th century, when the British authorities in the Colony of Sierra Leone adopted a hostile stance toward Temne-speaking chiefdoms during "trade wars" at that time, that a common "Temne" identity developed (Ijagbemi 1968). 8. Fernandes also tells us that both Temne- and Bullom-speakers recognized, in addition to the king of each village, a figure called the "lord of all," but that this lord had no more authority than other "lords" (Fernandes 1951 [1506-10]:80-83). 9. Writing of peoples further north, Alvares describes how herbalists give offerings to "the idols"—here Islamized as "jinn" (china)—in ritual requests to intercede for them with the witches who consume the sick and dying: if an illness is unknown to them and they will not run the risk of treating it, they obtain from the sick man something to give to the china Ijina], since they have persuaded the people that (when they are sick) witches are devouring them. By this trickery these priests make the man believe that they will go to the idol and will demand that it does not permit this to happen, by causing that person (? the witch) to feel regret and sorrow for the ill he has done, so that he then leaves the victim alone. But if the sick wretch 870 american ethnologist dies, they say that witches ate him up. No-one dies here, for to die is to be "eaten up." "They have eaten him" means "they have killed him." (1990(c.l615), pt. 1, ch. 7:10-11) 10. A similar account of the divinatory interrogation of the corpse and its litter is given forthe"Sonequei" (Soninke). This interrogation, however, may (literally) take a different path if the corpse propels the bearers of the litter past the "famous idol Manga Jata": instead of disclosing the human agents who "ate" the deceased, this reveals that the deceased was a witch killed by the "idol" (Alvares 1990[c.1615l, pt. 1, ch. 3:4). "Manga ]ata"—"King Jata"—presumably invokes Sundiata (or "Mari Djata" or "Djata"), the founder of the Malinke empire of ancient Mali that displaced the Soninke empire of ancient Ghana. Alvares laments the fact that Portugal's Atlantic enterprises were indirectly responsible for these practices, emphasizing his church's condemnation of particular methods of acquiring slaves (while it nevertheless condoned others), yet placing a more direct blame for these practices on what he called the "spite" of the heathen": Spite is (thus) one of the heads under which those who are sent to our Spanish lands [i.e., the Indies] are acquired, although this has been so strongly condemned in the bull promulgated by His Holiness in the year—on the subject of the purchase of Brasilians. [1990(c.1615), pt. 1, ch. 3:4; cf. pt. 1, ch. 5:2-3] 11. Piot's work on the slave trade among the Kabre in present-day Togo (1996a) raises an interesting question about possible interrelationships among the different slave-producing practices in Sierra Leone. In the 18th century, the Kabre dwelt in noncentralized communities on the watershed of the two great slave-trading states of Asante and Dahomey. In a context in which Kabre communities were subject to considerable slave raiding from these powerful states, Kabre men responded by selling their sisters' children in the marketplace—a strategy that gave them more control over the inevitable fact of slaving, and was ultimately "successful" in that it eventually replaced the raids. Although the sale of slaves through witchfinding and pawnship cannot be seen as parallel strategies for reducing raids and wars among Temne-speaking communities—for unlike the Kabre, those who bought slaves from Temne elites were not usually those who raided and waged wars against them—the issue of possible interrelationships among warfare, raiding, witchfinding, indebtedness, and pawnship on the Upper Guinea Coast deserves further investigation. 12. Many different forms of public divination are available. First, there are "moving vehicle" techniques (Dorjahn 1962:6) in which a single diviner uses an object charged with medicines that control the movements of one or several ritually treated bearers. As well as the "corpse-interrogation" techniques of "the funeral litter" lan-sanka) and "the coffin" (on-benthj) (divination methods similar to those that proliferated during the slave trade), moving vehicle methods also include "the broom" (arj-gba'b), "the pestle" (ka-ronp), and "the whip" (an-raket), all of which compel their bearers to beat an accused person who is thereby identified as the culprit. Second, spirit embodiments from certain male cult associations—ma-neke spirits of the m-Gbenlesociety in eastern Temneland (see Dorjahn 1959), the masked ka-yogbo spirit of the (originally Yoruba/Aku Krio) Oje society in western Temneland, and ao-Cbaogba'ni, the spirit of the Limba Gbangbani society along the Temne/Limba borderland—are extrahuman agents that disclose the guilt of witches and other wrongdoers through verbal interrogations, accusations, and pronouncements (Shaw 1985; 1991). Finally, there are also ordeals (rarely used now, to my knowledge, but vividly remembered), including "the ax" (ka-bap), whose blade is placed on the suspect's tongue after being heated in a fire until it glows; "boiling oil" (rm-kobe), into a pot of which the suspect reaches to bring out a piece of iron within; "the needle" (an-se'ni), which is threaded and passed through the skin along the suspect's forearm; and the infamous "red water" (ma-kon; cf. "the bark of macone," used for the same purpose in the early 1600s [Alvares 199O(c.1615), pt. 2, ch. 3:5]), a poison ordeal in which the suspect drinks liquid made from mashed sasswood bark. 13. In order to do so, they must pass a test known as "the concealment" (an-monk): a small object such as a coin is hidden and must be found through divination in order to demonstrate the capacity to discover what is concealed. 14. Whereas, like other oracles and ordeals, witchfinding divination during the slave trade was formerly "not evidence for reaching a legal judgment but the definition of a judgment itself" (Asad 1993:91), it has now become a source of "evidence" on which separate formal judgment is based. 15. In both the early 17th and the late 18th centuries, for instance, we are told that the identification of suspected witches began with names from the family of the afflicted or deceased person (Alvares 1990[c.1615), pt. 2, ch. 3:4; Matthews 1961 [1788]:123). Such accounts indicate that lineage politics informed cultural ideas about witchcraft in important ways. We are told by the same observers, moreover, that when a person was convicted as a witch, "a host of his children are enslaved" (Alvares 1990[c.1615], pt. 1, ch. 7:13), and that (in the 18th century, at least) "it frequently happens if the deceased were a great man, and the accused poor, not only he himself but his whole family are sold together" (Matthews 1961 [1788]:124). This implies, perhaps, that witchcraft was understood as shared among kin: it is, as we have seen, so understood today, although not as passed directly through the "blood" of kinship but as imparted by witches who recruit their relatives by feeding them human meat in dreams. To what extent might this "kinship theory of witchcraft"—entailing interpretations that authorized the enrichment of rulers through the sale of "whole families"—have been promoted by the presence of the slave trade? While we will probably never know the answers to this question, it is clear that neither representations of witchcraft nor the social relations behind witchcraft accusations in Sierra Leone in these centuries can be understood in isolation from the Atlantic system. production of witchcraft/witchcraft as production 871 16. In the historical sources I have examined, the gender of those accused of wilchcrafl is not usually specified (although Winterbottom 11803:139-143] gives a remarkable account of witchcraft accusations against newly inducted novices by officials of the women's Bondo society at the end of the 18th century: those who confessed or failed a medicine ordeal were sold into the Atlantic trade). Today, however, not only do understandings of witchcraft represent women as the majority of witches, but these understandings also resonate in telling ways with some strongly gendered consequences of the slave trade for Temne marriage. Although these representations cannot tell us whether more women than men were actually convicted of witchcraft during the slave-trade era, an investigation of how gendered labor needs in parts of the Americas and the Caribbean might have affected the demand for slaves of a particular gender in Sierra Leone could shed some light on this. Unfortunately, length constraints do not allow me to explore these issues here (but see Shaw n.d.). 17. It is often assumed that all big persons have witch capacities, not only in order to have achieved their wealth and power in the first place, but also to defend themselves from the enemies they must invariably accumulate (see also Bastian 1993; Ceschiere 1994; Rowlands and Warnier 1988). As long as big persons do not allow their capacities for predation and consumption to overwhelm their acts of patronage and generosity toward those in their own communities, these capacities may be admired (see Austen 1993:91). Such high-status witches often meet with approval because they are represented as protecting the community against the malignancy of others (see Austen 1993:91). The magically armed big person is correspondingly represented as able to strike down witches who "raid" the community both from outside and from within: in a world in which witchcraft is everywhere, I was often told, the only protection against witches is a witch who has more power. 18. As we have seen, those who kill from afar with witch guns (an-piokor o-seron) are the "big persons" of the witch-city. In examining the terrible preeminence given to these owners of invisible firearms, it is useful to outline the development of a new kind of leader in Sierra Leone during the 17th and 18th centuries: wealthy merchants who had become affluent and powerful through the slave trade (Rodney 1970:200). Often the descendants of European traders who had married into chiefly families, some of these men had traveled to Europe, been educated in England, and bore names such as Caulker, Cleveland, Tucker, and Rogers (see Fyfe 1964:100-103). These new leaders were able to accumulate considerable material wealth, which they used to acquire imported European commodities, including (and perhaps especially) firearms. The latter were not only commodities for which slaves could be exchanged, but were also a means of producing more slaves through the intensification of conflicts (Rodney 1970:176). When the slave trade was prohibited in the 19th century, it was replaced by what the British called the "legitimate trade." The new class of big men who controlled this trade in the Sierra Leone hinterland also controlled the flow of firearms and ammunition, obtaining these through the sale of forest products such as timber and palm oil (Ijagbemi 1968:205-210). When finally, at the end of the century, the British imposed a Protectorate on the hinterland of Sierra Leone, their colonial control was enforced (over a highly effective guerrilla resistance) by superior firepower. In Sierra Leone, then, as in other parts of Africa, European technologies of destruction have also been technologies of material wealth and predatory domination. As such, their intangible—but equally deadly—witch equivalents are a highly "appropriate technology" for the most powerful inhabitants of the witch-city. 19. Such recognition through images, rumors, and stories need not—as many have pointed out—take verbally discursive forms in order to "count" as historical consciousness. As Comaroff and Comaroff put it, "history lies in its representations, for representation is as much the making of history as it is consciousness speaking out" (1991:35-36). 20. Gable's criticism of this "dichotomizing history" is undermined, however, by his main argument. He concludes that contractual attitudes toward spirits and their "laws" in a Manjaco village further north along the Upper Guinea Coast in Guinea-Bissau are autonomous, sui generis creations because they do no! seem to have originated during a colonial period that began only in 1914. Yet by overlooking the several centuries of contractual commercial relationships that the slave trade era in this region entailed before colonialism supervened (Brooks 1993; Rodney 1970), his critique of analyses of "the colonization of consciousness" effectively returns us to an ahistorical anthropology. 21. The Colony of Sierra Leone, established in 1808, was the earliest Crown Colony in West Africa. Colonial rule was not imposed over the "hinterland," however, until the establishment of the Protectorate of Sierra Leone in 1896. references cited Alvares, Manuel 1990[c. 16151 Ethiopia Minor and a Geographical Account of the Province of Sierra Leone. P.E.H. Hair, trans. Department of History, University of Liverpool, mimeograph. Apter, Andrew 1993 Atinga Revisited: Yoruba Witchcraft and the Cocoa Economy, 1950-1951. In Modernity and its Malcontents, lean and ]ohn Comaroff, eds. Pp. 111-128. Chicago: University of Chicago Press. 872 american ethnologist Asad, Talal 1993 Genealogies of Religion: Discipline and Reasons of Power in Christianity and Islam. Baltimore, MD: Johns Hopkins University Press. Auslander, Mark 1993 "Open the Wombs!": The Symbolic Politics of Modern Ngoni Wrtchfinding. In Modernity and its Malcontents. Jean and John Comaroff, eds. Pp. 167-192. Chicago: University of Chicago Press. Austen, Ralph A. 1993 The Moral Economy of Witchcraft: An Essay in Comparative History. In Modernity and its Malcontents. Jean and John Comaroff, eds. Pp. 89-110. Chicago: University of Chicago Press. Bastian, Misty L 1990 Blood and Petrol: The Dangerous Eroticism of Roads and Transport in South-East Nigeria. Paper presented at the Annual Meeting of the African Studies Association, Baltimore, MD, November. 1993 "Bloodhounds Who Have No Friends": Witchcraft and Locality in the Nigerian Popular Press. In Modernity and its Malcontents. Jean and John Comaroff, eds. Pp. 129-166. Chicago: University of Chicago Press. Bayart, Jean-Francois 1993 The State in Africa: The Politics of the Belly. London: Verso. Bledsoe, Caroline H. 1980 Women and Marriage in Kpelle Society. Stanford, CA: Stanford University Press. Bourdieu, Pierre 1990 The Logic of Practice. Stanford, CA: Stanford University Press. Brooks, George E. 1993 Landlords and Strangers: Ecology, Society, and Trade in Western Africa, 1000-1630. Boulder, CO: Westview Press. Ciekawy, Diane 1992 Witchcraft Eradication as Political Process in Kilifi District, Kenya, 1955-1988. Ph.D. dissertation, Columbia University. Comaroff, lean, and John Comaroff 1991 Of Revelation and Revolution: Christianity, Colonialism, and Consciousness in South Africa, vol. 1. Chicago: University of Chicago Press. 1993 Introduction. In Modernity and its Malcontents: Ritual and Power in Postcolonial Africa. Jean and John Comaroff, eds. Pp. xi-xxxvii. Chicago: University of Chicago Press. Dirks, Nicholas B. 1992a Introduction. In Colonialism and Culture. Nicholas B. Dirks, ed. Pp. 1-25. Ann Arbor: University of Michigan Press. 1992b Castes of Mind. Representations 37:56-78. Donelha, Andre 1977{1625] Descricao da Serra Leoa e dos Rios de Guin£ do Cabo Verde (1625). (An account of Sierra Leone and the rivers of Guinea of Cape Verde [1625].) Avelino Teixera da Mota and P.E.H. Hair, eds and trans. Lisbon: Centra de Estudos de Cartografia Antiga. Dorjahn, V. R. 1959 The Organization and Functions of the Ragbenle Society of the Temne. Africa 29:156-170. 1962 Some Aspects of Temne Divination. Sierra Leone Bulletin of Religion 4:1-9. Douglas, Mary, ed. 1970 Witchcraft Confessions and Accusations. London: Tavistock. Evans-Pritchard, E. E. 1937 Witchcraft, Oracles and Magic among the Azande. Oxford: Clarendon Press. Ferme, Mariane C. 1992 "Hammocks Belong to Men, Stools to Women": Constructing and Contesting Gender Domains in a Mende Village (Sierra Leone, West Africa). Ph.D. dissertation, University of Chicago. Fernandes, Valentim 1951 [1506-1510] Description de la Cote Occidentale d'Afrique (Senegal au Cap de Monte, Archipels) par Valentim Fernandes (1506-1510). Th. Monod, A. Teixera da Mota, and R. Mauny, trans. Bissau, Portugal: Centra de Estudiosda Guin6 Portuguesa (No.11). Foster, George M. 1965 Peasant Society and the Image of the Limited Good. American Anthropologist 67:293-315. Fyfe, Christopher 1962 A History of Sierra Leone. London: Oxford University Press. 1964 Sierra Leone Inheritance. London/lbadan/Accra: Oxford University Press. Fyle, C. Magbaily 1988 History and Socio-Economic Development in Sierra Leone. Freetown: Sierra Leone Adult Education Association. Gable, Eric 1995 The Decolonization of Consciousness: Local Skeptics and the "Will to be Modern" in a West African Village. American Ethnologist 22:242-257. production of witchcraft/witchcraft as production 873 Ceschiere, Peter 1988 Sorcery and the State: Popular Modes of Action among the Maka of Southeast Cameroon. Critique of Anthropology 6:35-63. Ceschiere, Peter, with Cyprian Fisiy 1994 Domesticating Personal Violence: Witchcraft, Courts and Confessions in Cameroon. Africa 64:323-341. Ciddens, Anthony 1979 Central Problems in Social Theory. Action, Structure and Contradiction in Social Analysis. Berkeley: University of California Press. 1981 A Contemporary Critique of Historical Materialism. London: Macmillan. Gluckman, Max 1956 Custom and Conflict in Africa. Oxford: Blackwell. Gupta, Akhil, and James Ferguson 1992 Beyond "Culture": Space, Identity, and the Politics of Difference. Cultural Anthropology 7:6-23 Ijagbemi, E. Ade 1968 A History of the Temne in the Nineteenth Century. Ph.D. dissertation, University of Edinburgh. Jackson, Michael 1989 Paths toward a Clearing: Radical Empiricism and Ethnographic Inquiry. Bloomington: Indiana University Press. (ones, Adam 1981 Who were the Vai? Journal of African History 22:1 59-178. Kaspin, Deborah 1993 Chewa Visions and Revisions of Power: Transformations of the Nyau Dance in Central Malawi. In Modernity and its Malcontents. Jean and John Comaroff, eds. Pp. 34-57. Chicago: University of Chicago Press. Kearney, Michael 1995 The Local and the Global: The Anthropology of Globalization and Transnationalism. Annual Review of Anthropology 24:547-565. Laing, Major Alexander Gordon 1825 Travels in the Timmannee, Kooranko and Soolima Countries in Western Africa. London: John Murray. Latour, Bruno 1993 We Have Never Been Modern. Cambridge, MA: Harvard University Press. Littlejohn, James 1960a The Temne Ansasa. Sierra Leone Studies (n.s.) 14:32-35. 1960b The Temne House. Sierra Leone Studies (n.s.) 14:63-79. 1963 Temne Space. Anthropological Quarterly 36:1-17. "Look Gron" 1978 Le10 for Witchgun Blast? We Yone (Freetown, Sierra Leone), November 22:4, 10. MacCormack, Carol 1983 Human Leopards and Crocodiles: Political Meanings of Calegorical Anomalies. /nThe Ethnography of Cannibalism. Paula Brown and Donald Tuzin, eds. Pp. 51-60. Washington, DC: Society for Psychological Anthropology. MacGaffey, Wyatt 1986 Religion and Society in Central Africa: The Bakongo of Lower Zaire. Chicago: University of Chicago Press. Masquelier, Adeline 1992 Encounter with a Road Siren: Machines, Bodies, and Commodities in the Imagination of a Mawri Healer. Visual Anthropology Review 8:56-69. Matory, J. Lorand 1994 Sex and the Empire That Is No More. Minneapolis: University of Minnesota Press. Matthews, John 1961 [1788] A Voyage to the River Sierra-Leone, on the Coast of Africa; Containing an Account of the Trade and Productions of the Country, and of the Civil and Religious Customs and Manners of the People; In a Series of Letters to a Friend in England. London: B. White and Son. Mbembe, Achille 1992 Provisional Notes on the Postcolony. Africa 62:3-37. Meyer, Birgit 1995 "Delivered from the Powers of Darkness": Confessions about Satanic Riches in Christian Ghana. Africa 65:236-255. Miller, Daniel 1995 Introduction: Anthropology, Modernity and Consumption. In Worlds Apart: Modernity through the Prism of the Local. Daniel Miller, ed. Pp. 1-22. London and New York: Routledge. Miller, Joseph C. 1976 Kings and Kinsmen: Early Mbundu States in Angola. Oxford: Clarendon Press. 1988 Way of Death: Merchant Capitalism and the Angolan Slave Trade, 1730-1830. Madison: University of Wisconsin Press. 874 american ethnologist Mudimbe, V. Y. 1988 The Invention of Africa. Bloomington: Indiana University Press. Munn, Nancy D. 1990 Constructing Regional Worlds in Experience: Kula Exchange, Witchcraft and Cawan Local Events. Man (n.s.) 25:1-17. Peek, P. M., ed. 1991 African Divination Systems: Ways of Knowing. Bloomington: Indiana University Press. Piot, Charles 1996a Of Slaves and the Gift: Kabre Sale of Kin during the Era of the Slave Trade. Journal of African History 37:31 ^ t 9 . 1996b Circulating Identities in a West African Diaspora. Paper presented at the Annual Spring Meeting of the American Ethnological Society, San Juan, Puerto Rico, April. Public Records Office (PRO) 1892-99 Protectorate Ordinances of Sierra Leone—CO/269/5. Kew, London. 1899 Chalmer's Report on the Rebellion—CO/879/54, Appendix I: Oral Evidence. J. C. E. Parkes, Secretary for Native Affairs, 29 July 1898. Kew, London. Rankin, F. Harrison 1836 The White Man's Crave: A Visit to Sierra Leone in 1834, 2. London: Richard Bentley. Richards, Paul 1986 Coping with Hunger: Hazard and Experiment in an African Rice-Farming System. London: Allen and Unwin. 1995 Chimpanzees, Diamonds and War: The Discourses of Global Environmental Change and Local Violence on the Liberia-Sierra Leone Border. In The Future of Anthropological Knowledge. Henrietta L. Moore, ed. Pp. 139-155. London and New York: Routledge. 1996 Fighting for the Rain Forest: War, Youth and Resources in Sierra Leone. Oxford: James Currey; Portsmouth, NH: Heinemann. Rodney, Walter 1970 A History of the Upper Guinea Coast 1545 to 1800. London: Oxford University Press. Rowlands, Michael, and Jean-Pierre Warnier 1988 Sorcery, Power and the Modern State in Cameroon. Man (n.s.)23:118-132. Sahlins, Marshall 1988 Cosmologies of Capitalism: The Trans-Pacific Sector of "the World System." Proceedings of the British Academy 74:1-51. Sayers, E. F. 1927 Notes on the Clan or Family Names Common in the Area Inhabited by Temne-Speaking People. Sierra Leone Studies (o.s.) 5:14-109. Shaw, Rosalind 1985 Gender and the Structuring of Reality in Temne Divination. Africa 55:286-303. 1991 • Splitting Truths from Darkness: Epistemological Aspects of Temne Divination. In African Divination Systems: Ways of Knowing. Philip Peek, ed. Pp. 137-152. Bloomington: Indiana University Press. 1992 Dreaming as Accomplishment: Power, the Individual and Temne Divination. In Dreaming, Religion and Society in Africa. M. C. Jedrej and Rosalind Shaw, eds. Pp. 36-54. Leiden: E. J. Brill. 1996 The Politician and the Diviner: "Big Men" and the Ritual Consumption of Power in Sierra Leone. Journal of Religion in Africa 26:30-55. In press "Tok Af, Lef Af": A Political Economy of Temne Techniques of Secrecy and Self. In African Philosophy and Critical Inquiry. Ivan Karp and D. A. Masolo, eds. Bloomington: Indiana University Press, n.d. The Dangers of Temne Divination: Ritual Memories of the Slave Trade in West Africa. University of Chicago Press, forthcoming. South Carolina ETV 1992 Family across the Sea. Video. Taussig, Michael 1992 Culture of Terror—Space of Death: Roger Casement's Putumayo Report and the Explanation of Torture. In Colonialism and Culture. Nicholas B. Dirks, ed. Pp. 135-173. Ann Arbor: University of Michigan Press. Turay, Abdul Karim 1971 Loanwords in Temne: A Study of the Sources and Processes of Lexical Borrowing in a Sierra Leonean Language. Ph.D. dissertation, University of London (SOAS). Turner, Victor W. 1957 Schism and Continuity in an African Society: A Study of Ndembu Village Life. Manchester: Manchester University Press. Weiss, Brad 1996 The Making and Unmaking of the Haya Lived World: Consumption, Commoditization, and Everyday Practice. Durham, NC: Duke University Press. White, Luise 1990 Bodily Fluids and Usufruct: Controlling Property in Nairobi. Canadian Journal of African Studies 24:418-438. production of witchcraft/witchcraft as production 875 Wilmsen, Edwin N. 1989 Land Filled with Flies: A Political Economy of the Kalahari. Chicago: University of Chicago Press. Winterbottom, Thomas 1969 [ 1803] An Account of the Native Africans in the Neighbourhood of Sierra Leone, to Which is added an Account of the Present State of Medicine among Them, 1. London: Frank Cass. submitted May 31, 1996 accepted November 27, 1996 Latino Cultural Citizenship Claiming Identity, Space, and Rights EDITED BY W I L L I A M V . FLORES CULTURAL CITIZENSHIP AND R I N A B E N M A Y O R Written by Renato Rosaldo LATINO Ana L Juarbe Richard R. Flores Rosa M. Torruellas Raymond Rocco William V. Flores Blanca G. SUvestrini Rina Benmayor CLAIMINC IDCHTITT Ifllt A N D I11ITI A vivid, innovative portrait of Latino culture, based on ethnographic work in San Antonio, Los Angeles, WIIUAMFIOIEI » UNA MNMAYtl New York, San Jose, and Watsonville, California. The authors explore Latino "cultural citizenship" —the use of cultural expression to claim political hardcover $27.50 rights while maintaining a vibrant local identity. B E A C O N 876 american ethnologist P R E S S h t t p : / / w w w . beacon, a

RELATED PAPERS

RELATED TOPICS

Log In

The production of witchcraft/witchcraft as production: memory, modernity, and the slave trade in Sierra Leone

The production of witchcraft/witchcraft as production: memory, modernity, and the slave trade in Sierra Leone

Related Papers

RELATED PAPERS

RELATED TOPICS