International African Institute

Transcription

1 International African Institute Slavery, Possession and History: The Construction of the Self among Slave Descendants in the Sudan Author(s): G. P. Makris Source: Africa: Journal of the International African Institute, Vol. 66, No. 2 (1996), pp Published by: Cambridge University Press on behalf of the International African Institute Stable URL: Accessed: 11/10/ :16 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Cambridge University Press and International African Institute are collaborating with JSTOR to digitize, preserve and extend access to Africa: Journal of the International African Institute.

2 Africa 66 (2), 1996 SLAVERY, POSSESSION AND HISTORY: THE CONSTRUCTION OF THE SELF AMONG SLAVE DESCENDANTS IN THE SUDAN G. P. Makris Scene I. 'Who are they?' I asked Wad al-gibel, a low cult-officiant with whom I had become friends after countless bouts of drinking low-quality 'araqi.'they are the leaders of tumbura in the area', he whispered absentmindedly. The still early morning air was too hot even for him. Well, that was all I needed. If Wad al-gibel were in his early seventies, I thought, these people should belong to ancient history. That was all I needed. The devotees of the tumbura spirit possession cult whom I was trying to meet proved to be a bunch of old men of disreputable and dirty appearance who-in addition-did not seem to like me. A feeling of unease gripped my stomach. I did not even know where I was. Somewhere in Khartoum North. Evidently the whole area of uniform, monotonouslums had been reduced to rubble by the floods of last August. We were sitting under a makeshift tent of cardboard boxes, cloth, garbage and rusty iron sheets. The flies, sweeping down on my teacup, buzzed incessantly. How could they live in such dirt? Sambu, the old tumbura officiant who had brought me there with Wad al- Gibel, his assistant, interrupted my train of thought. 'The women will not come for another two hours. We will not hit the drum before supper.' I looked at his maimed leg, the dirty jallabiyya and the protruding red eyes. I disliked him intensely. 'Sleep, sleep,' he continued. 'I will call you.' I slept. I was awakened by Wad al-gibel. It was midday. 'They do not want you here,' he said quietly. I did not comprehend what he was saying. I looked at him, still half asleep. 'Marfssa Moya talked to the old ones. They do not like you. I will put you on the bus to Khartoum.' I looked for Sambu. He was not around. As we were leaving the place, shaykha Marlssa Moya, the most famous tumbura leader in Khartoum North, approached us. She was a very old, black, thin, diminutive woman. She smelled. Her yellowish white hair contrasted with her patched, faded black tob. No shoes. No teeth. She said something I did not catch; she laughed. Wad al-gibel was looking impatient. At the bus stop he said, 'The tumbura people do not like strangers'. Disgusted by my failure and their intransigence, I left. Scene II. Two months later, having finally established contact with the tumbura groups of southern Umm Badda, outside Omdurman: 'Why do you want to go to Umm Badda?' asked the taxi driver. 'There is a tumbura ceremony going on today. I work for the University.' 'This tumbura is the dance of the Nubas, eh?' 'There are some Nubas there,' I agreed half-heartedly. 'Why do you want to go and see the slaves? You are a Khawaja. I will take you to see the Darawish in Hamad al-nfl.' Mildly irritated, I laughed. 'No, no; it doesn't matter. Go to Umm Badda.' An hour later: 'Grandmother!'

3 160 TUMBURA IN SUDAN 'What?' shaykha Halima, a tumbura leader, replied. 'Who are the tumbura people? where do they come from?' 'Markfsh! Can't you hear properly? I've told you already,' she replied half angrily. Then she added, more indulgently, 'They are from the country, from the ninety-nine mountains.' Twenty minutes later: 'Who are the tumbura people, shaykha Nfira?' 'They are the good Muslims, the people of Bilil, the muadhdhin of the Prophet.' Scene III. A week later in Sambu's house in Fitfhab: 'Who are the tumbura people, Wad al-gibel?' 'They are the Sfdani, the original people of the land (nds dsli) who were enslaved by the Arabs.' MAKING SENSE OF POSSESSION I went to Khartoum as a novice field-worker in 1988 to study tumbura, a spirit possession cult related to zar bore, the more widely known cult of the area. Constantinides (1972, 1991), Kenyon (1991a,b), Samia al-hadi al- Nagar (1975, 1980, 1987), Trimingham (1983) and Zenkovsky (1950) have argued that tumbura is associated with descendants of black slaves. Ostensibly, then, the taxi driver was not far from the truth. Yet, although vilified by outsiders as slaves, pagans, drunkards, thieves and prostitutes, the tumbura devotees, I was to find as I got to know them better, presented themselves in a positive, dignified and circumspect manner. Still, to outsiders they remained elusive and suspect in the way they presented themselves as the autochthonous people of the land. But what did they really mean by that? What was the relation between these conflicting identities and tumbura? Did tumbura represent a single unified practice or was it experienced by its devotees in diverse ways? What role did possession play in the construction of the devotees' everyday reality? Answers to these questions are inevitably inflected and delimited by the existing theoretical approaches to spirit possession. It is impossible to do justice here to the richness of possession literature, as the themes it explores are many and diverse.' Identifying major trends, we see that some researchers examine possession predominantly as a therapeutic practice, or emphasise its strategic use and the benefits that the possessed enjoy (Kehoe and Giletti, 1981; Kennedy, 1967; Lewis, 1966, 1971; Messing, 1958; Ward, 1989). Other researchers adopt more interpretive and context-sensitive approaches that bring together medicine and religion; they variously emphasise the role of possession in the construction of meaning for the devotees; see possession as a mechanism of urban adaptation; regard it as part of subordinate (mainly female) discourses; or link it with the wider problem of the constitution of selfhood, drawing attention to its aesthetics as theatre, allegory and historical metacommentary (Ardener, 1972; Boddy, 1989; Brown, 1991; Constantinides, 1972, 1977, 1979; Crapanzano, 1973, 1977, 1980; Kapferer, 1983; Kramer, 1993; Lambek, 1981; Lewis, 1977, 1971/1989; Obeyesekere, 1984; Rasmussen, 1995). The problem with most theories of the first type is that they often reduce the richness of possession by disregarding its allegorical nature; rationalise it through the imposition of a largely assumed intentionality; and unwittingly

4 TUMBURA IN SUDAN 161 adopt views propagated by the (mostly male) bearers of native dominant discourses.2 Those of the second type occasionally present another kind of shortcoming. By focusing intensely on the intellectual and cognitive aspects of possession they can undervalue the social dimension of possession cults and the actors' positioning within their organisational structure.3 The weakness, though, that concerns me here and which can be seen in both types of approach is that, although they associate possession with power struggles located in relations of inequality, they often say little about how those relations change in time and how possession cults can transform and simultaneously, be transformed by them.4 Certainly, there are exceptions, but on the whole possession cults until recently have been examined as if they were practices that neither affect nor are affected by historical change. In this article I shall attempt to introduce this rather neglected historical dimension to the analysis of the tumbura spirit possession cult. In particular, I shall study the relationship between tumbura and the stigmatised slaverelated identity of its practitioners by investigating how cult participation has been enabling the cult devotees to articulate a positive self-identity in the context of colonial and post-colonial Sudan. In this view tumbura is seen as providing an alternative record of history. SETTING THE STAGE: PRACTICES AND IDEOLOGIES The era of slavery The 1898 Anglo-Egyptian reconquest of the Sudan found the country devastated. The Turco-Egyptian occupation ( ) and a harsh Islamic state under the Mahdists ( ) had led to the collapse of the structure of the local societies. Central for both regimes was the enslavement of pagan Africans from the Southern and Western Sudan by Northern Sudanese slave traders, with the direct or indirect assistance of the authorities. Most of the slaves were engaged in agriculture, domestic service and the military. Slavery had been present in the region for many centuries and until the Turco-Egyptian invasion it had a rather 'benevolent' character in accordance with the injunctions of Islamic law (Hargey, 1981: 16 ff.). After 1821, however, the political developments marginalised the Islamic 'paternal' and assimilative attitudes and reduced the status of slaves considerably (Hargey, 1981: 22; Hill, 1959; Holt, 1970; Holt and Daly, 1988; Stiansen, 1993). The nineteenth-century slaving frontier was defined by the actors in terms of Islamic versus non-islamic pedigree, brown versus black colour, Arab versus non-arab descent (O'Fahey, 1982: 75). In the present context the term 'Arab' should be understood not in a genealogical sense but as an ideological pole that joined to the Arabian heartlands of Islam the Arabised Northern Sudanese peoples,6 who were Muslims and, significantly, kept slaves. Among these peoples there were also several others who did not claim Arab descent although they were more or less engaged in a gradual process of Arabisation. Specific tribal affiliations as well as competition over resources certainly existed between all these diverse groups; however, there was also present a feeling of belonging to one imagined community7 of Muslim freemen. If this community could be called Arab, it is a matter of

5 162 TUMBURA IN SUDAN interpretation as the self-described Arab populations occupied increasingly higher positions of power at the political as well as the cultural level. What gave this imagined community of Muslim freemen an aura of permanence and defined it as the realm of freedom, humanity and Islamic tradition was the construction of its antithetical image: an equally imagined community of enslavable pagan black Africans with no religion, history or descent who were living on the other side of the frontier. They were collectively seen as 'speaking animals' (O'Fahey, 1983: 244; Warburg, 1992: 159, n. 68) and were reduced to the level of 'mere commodities... classified in the legal sense with livestock' (Hargey, 1981: 22-3).8 In actual practice this ostensibly absolute distinction between the two imagined communities was subject to a process of constant negotiation that governed the shifting of their boundary without, however, dissolving it. Central to this negotiation was the Islamisation and Arabisation of the slaves, both of which processes were perceived differently by the two sides. For the slave owners the conversion of the slaves to Islam was not only a conscious action of introducing them to religion and culture but a practical necessity as well. Among other things, the slaves slaughtered animals, prepared meals, entered into sexual relations with their masters and were enrolled in armies under Muslim officers. Had they remained pagan they could not have engaged in such activities. But, more than that, as Cooper suggests of the East African coast in general (1981: 290), the conversion of the slaves to Islam effected the internalisation of the religious basis of their subordinate status. Without denying, then, that Islam brought the two sides closer, we must also see it as a vehicle of cultural hegemony that ensured the subordination of the enslaved populations. This picture is completed when we consider the Arabisation of the slaves that was intrinsically related to their Islamisation and entailed the adoption of Arab customs, Arab codes of behaviour and the Arabic language. In the eyes of the slave owners Arabisation represented an improvement that could only rarely reach equality. What stood in the way were physical characteristics, colour and descent. To paraphrase Cooper (1981: 291), the slaves could become Muslims but they could not become as good Muslims as their masters. What they lacked was descent; they were not members of the imagined community of Muslim freemen. Certainly there were some who more or less 'crossed' to the other side, but for the majority the stigma of slave descent was carried from generation to generation. As Hayder Ibrahim (1979: 81) argues, 'to be a slave (sc. ex-slave or slave descendant) or to be free continues to be the basic criterion for the stratification of the village'.9 For the slaves, conversion to Islam opened avenues of limited social and economic advancement in wider society on a more equal basis with the freemen (Hargey, 1981: 23-6), and offered them the only viable system of supernatural protection in their hostile new environment.'0 However, this does not mean that the slaves were ready to forget, disregard or belittle their own cultural past as a whole. Their evolving Islamic/Arabised tradition was different from that of their masters and included practices and ideas such as ecstatic dances, singing and spirit possession cults like tumbura that went back to their native cultures (cf. Caplan, 1975; Cooper, 1981: 290; Glassman, 1991; Lewis, 1986: 94 ff.).

6 TUMBURA IN SUDAN 163 Such divergent interpretations of Islamisation and Arabisation illustrate the porous nature of the slaving frontier that lessened the distance between the two imagined communities, while at the same time they highlight the frontier's very existence as a boundary that kept the two sides apart. The era of emancipation At the time of the 1898 reconquesthe slaves constituted 25 per cent of the population of the Northern Sudan. Facing such a stark reality, the colonial administrators abolished slavery in a gradual manner. Their principal fear was that unconditionally and suddenly freeing vast numbers of slaves could lead to the economic disintegration of agriculture and to a rebellion of the slave-holding populations, who, after all, 'had bought their slaves and paid for them' (Jackson, 1955: 93-4). In this way the administration 'conferred de facto legitimacy on the prevailing social code' (Hargey, 1981: 74), and, although the legal position of the freed slaves improved, the traditionally negative attitudes of the politically and economically dominant Northern Sudanese towards them and their descendants remained, from this point of view, intact.11 During the fifty-eight years of colonialism the imagined community of Muslim freemen outlined above became more discernibly Arab in character as the self-described Arab populations consolidated their already exalted position in it. Tribal loyalties continued to exist, but now more than ever can we talk more precisely about an Arab elite that included people from most of the Arab riverain tribes.12 This was due to internal reasons associated with the emergence in 1920 of a version of Sudanese nationalism, moulded by political and other aspirations of the Arab riverain people, as well as by the existence of an unspoken pact between the colonialists and these people.13 Indeed, the colonial government was greatly involved in these developments, since right from the beginning it had classified the population of the country into natives or Arabs and Sudanese, i.e. those 'who were in a state of slavery or considered as such by the natives [i.e. Arabs]' (Hargey, 1981: 102).14 The advent of independence in 1956 signalled the transfer of authority directly into the hands of these Arabs, bringing to power the same populations who sixty years earlier had been engaged in the slave trade (Warburg, 1978: 245). Since that day Arabisation has been a policy of most Sudanese governments.15 One would thus have expected the divide between them and the slave descendants and other subordinate Muslim populations who lived in the North to assume its previous violent character. However, that did not happen. Two reasons can be cited. First, since the 1950s all slave descendants have belonged to generations which were born in the North; consequently they are far more Arabised than their parents and grandparents. Second, in 1956 the non-muslim Southerners, the archetypical slaves of the past, who until that time had been kept by the British colonialists outside all political deliberations at a national level, 6appeared as visible actors on the political scene.17 Since that year the Southerners have been perceived, and to an extent have perceived themselves, as a single 'monolithic' group positioned against the Northerners in a civil war which, with an interlude of peace during President Numayrf's era, continues today.18 Such being the case, both the Arab Northerners and the Arabised slave descendants have

7 164 TUMBURA IN SUDAN moved closer to each other-though the latter are still seen as inferior by the former-because of the emergence of this third, non-muslim, marginal category. The changing position of the slave descendants in the social formation and their realignment with their erstwhile masters vis-a-vis a new marginal category reveals how dynamic are perceptions of identity in the Sudan. One example of such dynamism can be found in the views of a group of slave descendants, the devotees of the tumbura spirit possession cult. THE MAIN ACTORS: THE ARABS AND THE SUDANI The tumbura devotees have a situational identity the transformations of which follow local political developments. My analysis is based on the views of elderly cult devotees, both male and female, as they were presented to me in in the area of Greater Khartoum. For these people tumbura is an Islamic healing cult intimately associated with themselves as members of a distinct social category, the Sudani people. As they say, 'Tumbura is a Sudani thing... God gave the Sdaiinf only tumbura.' Moreover the older devotees see themselves as the last generation of tumbura devotees and, consequently, as the last true Suidanf. For their part, the younger cult devotees, the majority of whom are women, do not entirely subscribe to these views and, without dismissing them in any obvious way, put them at the edge of their rhetoric. As we shall see, for them tumbura is not so much a Sudani Islamic healing cult as an Islamic healing cult. This shift of emphasis is of great importance and has brought tumbura closer than ever to zar bore, a similar possession cult that is particularly popular among the Arab Northerners. For me this multivocality shows that there is no real or authentic tumbura to be described, only different yet closely interrelated interpretations which change with time-and with locality. Consequently even this schematic distinction between two types of interpretations held by older and younger devotees respectively could be misleading, as the former type logically represents a 1980s crystallisation of other (diverse) views previously held by cult members. Therefore to privilege what I would call the interpretation of the older devotees and not that of the younger ones is not to imply that the former constitutes a fixed and absolute yardstick and the latter a deviation from or a distortion of it. I simply find that following their interpretation I am able to make better sense of tumbura's presence in the last hundred years or so rather than if I had followed any other. After all, in an almost selfcontradictory manner, everybody seemed to agree that what the elders were holding so dear represented the Old Truth (al-.haqq al-qadfm) that had been handed down to them by previous generations. That said, let me now turn to the ethnographic account itself. According to the elderly devotees, the people of tumbura (nds tumbura) are the Sudanin. The term refers to two different categories of people. The first concerns the descendants of the 'original people' (nds asl) of the country who were enslaved by the Arabs in the nineteenth century:

8 TUMBURA IN SUDAN the Sudani are from the South and the West, the lands of the Fartlt, the Kara, the Banda, the Nyam-Nyam. The Arabs went there when the Turk [Turco-Egyptians] were still in the country and took the Sudanin as slaves... They were selling them in the market as if they were goats.19 The second category of people concerns mainly the descendants of early Muslim migrants from the Nuba hills, the South and the West who came to the North during the first decades of the century. Although they have never been enslaved themselves, just like the ex-slaves and their descendants these people do not belong to the imagined community of the predominantly Arab Northerners and, consequently, they occupy similar subordinate positions within the socio-economic formation of the Northern Sudan. It is precisely this feeling of shared misery caused by slavery and general subordination that makes these disparate groups of people an identifiable entity in the context of tumbura discourse. What all this amounts to in terms of the previous discussion of Sudanese history is an inverted conceptualisation of the dichotomy between the two imagined communities I have described above:20 the slaves and their descendants, as well as other non-arab Muslim subalterns are described as members of a victimised imagined Sudanf community of 'original people', while the freeborn Northerners are seen as evil enslavers, members of an equally undifferentiated imagined Arab category. In this view the tumbura discourse seems to provide an alternative reading of the history of slavery and to facilitate the construction of a positive self-identity. However, this is not without its problems, as such a reading does not adequately reflect reality. To start with, the Sudanf identity is the result of actual or ideological enslavement: a free Nuba is a Nuba; an enslaved Nuba becomes a Sudanf. For that reason the cult devotees, especially the younger ones who were born and raised in the North, use it almost exclusively in the context of the cult.21 Outside this context, they adopt the Arab identities of the groups which enslaved their forefathers, claim the identity of the dominant groups in the area where they were born, or use locational terms of identification.22 This situation is clearly reflected in the fact that the social organisation and descent system of the Sudani cult devotees have been modelled on those of the Arabs. This, however, is not recognised by the tumbura discourse, which emphasises the rift between the Arabs and the Sudanf and ignores the negotiation processes concerning identity that have always been present between the two imagined communities. In similar vein the tumbura discourse presents both the Arabs and the Sudani as endogamous categories. This is not true. On the contrary, there has been a considerable number of marriages of freemen to enslaved women, although the opposite has been rarer. Lastly, the tumbura discourse describes a social reality that is not there any more. However alive, the distinction between freemen and slave descendants pales in political importance before the distinction between Northerners and Southerners, one that is conspicuously absent from tumbura discourse although it is really at the centre of contemporary political struggle. 165

9 166 TUMBURA IN SUDAN Such problems or obscurities are of the greatest importance to tumbura, for two reasons. First, they show that Stidani self-identity has always been predicated upon the Arab dominant identity, something that makes it structurally predisposed towards its own destruction. This may happen gradually or all at once when the objective conditions make all or some of the Sudani, especially the younger generations, able and willing to discard their Sudann-ness completely in favour of a viable (i.e. relatively acceptable) Arab identity. Second, they show that tumbura discourse does not reflect contemporary political reality and as such has limited relevance to the life of present-day subalterns. These either stay away from tumbura because they are non-muslim Southerners-the Sudanl are by definition Muslims-or they participate in the cult but advance their own novel understanding of it and its role in their lives. I will return to this point in the last section of the article. THE SHOW: ZAR BORE AND TUMBURA Before examining the ways in which this positive Sudanf self-identity is practically articulated in the context of tumbura I should briefly describe the wider ceremonial context that includes tumbura and zdr bore, the two major possession practices which are associated with the cult of the zdr spirit. As part of a single zdr cult complex, both practices share many structural and aesthetic characteristics. However, they have important differences as well, especially in relation to the sociological profile of each practice and the manner in which the zdr spirit is understood. It is the differences rather than the similarities that are stressed by their respective devotees. Zar bore is a predominantly female practice well embedded in the context of Northern Sudanese culture (Ahmed al-shahi, 1984; Boddy, 1989; Constantinides, 1972, 1991; Kenyon, 1991a,b; Lewis, 1989; Lewis et al., 1991; Samia al-hadi al-nagar, 1975, 1987).2 It is 'at once a healing rite and a parodical means to domesticate male and alien powers, an ambiguous metacommentary on local morality, and a history and anthropology of life in colonial and post-colonial Sudan' (Boddy, 1994: 417). The bore pantheon includes distinct spirits which are representatives of cultures other than that of the possessed (Boddy, 1989: 254, 301; Constantinides, 1977: 72; Kenyon, 1991b: 107). Possession by one or more of these spirits is expressed in the form of illness that is dealt with through the celebration of a ceremony and the satisfaction of the spirits' demands. These are usually expressed in terms of ritual obligations (sacrifice) and specific items. However, the patient is invariably attacked by the spirits repeatedly and has to go through the same steps every time, thus rebuilding anew, but always on a temporary basis, the positive relationship between the two. Different authors emphasise different aspects of bore in their attempts to analyse it,. Thus Lewis (1986, 1989) sees bore as a peripheral cult through which female devotees periodically gain material and moral support, while some graduate into leaders (shaykha) of independent cult groups; Constantinides (1977, 1979) analyses it as part of a female discourse centred on problems of female circumcision and infertility; while Boddy (1989) regards bore possession performances as allegorical texts that help

10 TUMBURA IN SUDAN 167 female patients to overcome their 'cultural overdetermination' and cultivate their self-awareness. Tumbura, on the other hand, has not received much attention in the literature; besides Seligman (1914, RAI), Zenkovsky (1950), and Kriss and Kriss-Heinrich (1960, 1962), only Kenyon (1991a,b) has offered a more recent eye-witness description of ceremonies.24 The cult has always been less widespread than bore. In the Turco-Egyptian occupation it was popular among male and female slaves, while it was suppressed during the Mahdist period as non-islamic. After 1898 tumbura re-emerged among the freed slaves, especially those who were serving in the Sudanese battalions of the Anglo-Egyptian army and their womenfolk,25 and took root mainly in the impoverished neighbourhoods of the urban centres and the surrounding shanty towns. In the 1960s it started to decline, but since the 1980s, at least in the area of Greater Khartoum, it has entered a period of restructuring that has brought it closer to the other practice of the zar cult complex, bore, and attracted again new devotees, particularly women.26 The reasons behind this are associated with the Arabisation of the tumbura people and are discussed in the last section of this article. In tumbura, zdr is not a generic term which covers many distinct entities, as in bore, but a single spirit that causes a wide range of illness.27 These are treated in the house of the female group leader called shaykha with the employment of traditional medicines and incense. It is only after the elimination of the symptoms and the full recovery of the patient that the latter is initiated into the cult. When this happens, he/she is shielded from any further relapses. In contrasthen with bore, tumbura ceremonies are not therapeutic in themselves. Rather, a three-day ceremony is celebrated after the patient's recovery as thanksgiving; a seven-day one marking his/her initiation into the cult follows, sometimes after a considerable period of time while other ceremonies are held annually to celebrate Islamic religious festivals. In all types of ceremonies the spirit 'descends upon' the devotees, assuming a number of forms which are intimately related to their historical experience. Contrary to bore, these forms are seen not as representatives of foreign cultures but as part of the devotees' collective Sudani self which emerges through their ecstatic dance. I will return to this point later. At the structuralevel tumbura also differs from bore. There each group functions independently, whereas in tumbura a small number of groups from the same area (and their respective shaykhas) come under a sanjak, the head of the local male hierarchy. During all the rituals the sanjak, accompanied by a rababa (literally, 'lyre'), sings the cult songs that invite the spirit to manifest itself through its hosts. Without him the spirit cannot really be approached in an ordered way. Ritually, then, the sanjak is indispensable, though generally his power is less pronounced today than it was in the past, when he was also involved in organisational matters.28 The crowning moment of tumbura is the sacrifice, during the initiation ceremony, of a sheep to honour shaykh Abd al-qadir al-jflanf, the founder of the Qadiriyya Sufi brotherhood, one of the major Sufi orders in the Sudan. The novice for whom the ceremony is held drinks some of the animal's blood, thus becoming a full member of the cult. At that moment the shaykh and the tumbura spirit are almost fused.

11 168 TUMBURA IN SUDAN As the above discussion suggests, tumbura is different from bore in many respects, especially since, after his/her initiation, the patient suffers no further relapses, and since the forms which the spirit assumes during the rituals are seen as part of the devotee's Sudanl self rather than as alien intruders. In the remaining sections of this article I will concentrate on the analysis of these features of the cult, arguing that tumbura must be seen as an aspect of the Siidian counter-discourse I have presented in the previous section, an invented Islamic tradition that belongs to what Scott (1990) calls 'infra-red politics', a kind of 'muted' but powerful resistance to the discourse of the predominantly Arab Northerners. THE MORAL: DENYING SUB-HUMANITY Identifying themselves as Suidann, the original people, the tumbura devotees turn the very fact of their enslavement by the dominant Arabs on its head, with the Sudadnf signifying not the sub-human slaves but the lawful owners of the land who have been wronged by the Arabs. At the level of practice this transformation is effected through their initiation into tumbura. The cult, which operates as an 'officialising accommodational strategy'29 and as a response to the lived experience of victimisation exemplified in the idiom of illness, is the field within which the Sidani devotees assert their humanity by affirming three elements which the Arabs seem to deny to them when they regard them as sub-human, namely religion, descent and history. Acquiring religion: tumbura as part of the Sufi tradition The Sufi brotherhoods have traditionally provided many Northern Sudanese with a religious, moral and political identity.30 On the other hand, this is not so for the slave descendants, who have not been particularly welcomed to join them.31 In many respects this vacuum is filled by tumbura, which is presented by its devotees as an Islamic institution similar to a brotherhood.32 I shall argue that this assertion is based (1) on the fact that the relationship between the spirit and its host in tumbura is modelled on that between a Sufi saint and his followers; and (2) on the similarities in devotional life and ritual structure between tumbura and the brotherhoods.33 The relationship between the spirit and its host in tumbura is determined by the fact that there are no relapses of the disease. Initiation into tumbura transforms the spirit into a benevolent force that secures good health, prosperity and general well-being. On his/her part the devotee must participate in the three annual Islamic festivals of Rajabiyya, Mawlid al- Nabf and (Id al-fitr, which are also celebrated by all Sufi brotherhoods, and attend the ceremonies performed by other devotees. Having celebrated the initiation ceremony, the devotee is not obliged to finance any more private ceremonies. If however a devotee breaks away from the cult the spirit will withhold its support, something that can cause illness, bad luck or death. However, such events are not regarded as a result of the spirit's will and, consequently, cannot be treated by the cult. They are ordinary misfortunes which would not have occurred if the relationship between the spirit and its host had remained close. This mechanism reminds us strongly of the withholding of a saint's

12 TUMBURA IN SUDAN 169 blessing (baraka) from an estranged follower. There, as in tumbura, one's active relationship with a saint (through initiation into a Sufi brotherhood) secures a continuous flow of the saint's baraka into one's life. Failure on the part of the supplicant to keep the relationship alive does not mean that any forthcoming afflictions have been caused by the saint. In that respect tumbura follows the Sufi tradition and the cult of the saints with the spirit considered to be a source of blessing just as the saint is considered to be a source of baraka. The similarity becomes clearer when we observe that the central ritual act of the initiation ceremony is the slaughtering of a sheep to honour shaykh Abd al-qadir al-jflanf, founder of the Qadiriyya brotherhood. Such is the importance of the shaykh in the context of tumbura that devotees often refer to him and to the spirit almost interchangeably, while there are many who actually identify the cult with the Qadiriyya brotherhood (altumbura di zdta Qadiriyya). Moving on to the devotional and ritual structural similarities between the cult and the Sufi brotherhoods, we see that tumbura follows an annual devotional cycle analogous to that of the brotherhoods. In this cycle, emphasis is given to the celebration of the three annual ceremonies which keep open the channel between the group and the source of blessing: in the case of tumbura, between the devotees and the spirit, and, in the case of the brotherhoods, between the followers and their saint. Also the initiation ceremony seems to have been modelled partly upon the hadra ceremony of the Khatmiyya brotherhood. Among other things, they both start with a procession where similar flags and banners are displayed; the songs of the tumbura procession resemble those of the brotherhood, mentioning the names of famous Sufis such as Ahmad al-rifati, Ahmad al-tijanf, Muhammad al-hasan b. Muhammad 'Uthman al-mfrghanf and his grandson, Sayyid All al-mfrghanfi34 and several ritual aspects which follow the tumbura and the Khatmiyya processions closely resemble each other. Such aspects include the use of incense for the fumigation of special items and of those present, the recitation of the Fatiha (the opening chapter of the Quran), the chanting of the supplication Shai'un li,llah ('Give something, in the name of God'), parts of the dhikr such as the Tahll ('There is no God but God') and Daym-Daym Allah ('God is eternal'), as well as the type and timing of food served.35 The similarities between the cult and the Sufi brotherhoods enable Sudanin devotees to argue that tumbura is an Islamic institution, and, consequently, to claim the Islamic credentials denied them by the Arabs, thus positioning themselves in the same social and moral universe as the latter. As they say, 'The Arabs are the people of the Book [Quran] and see tumbura as nonsense. This is not true. Tumbura is Islam. It was given by the Prophet to Bilal for the recovery of all the people' (lil-shafa, kul al-nds). Acquiring descent: tumbura ceremonies and the sons of Adam It is common in possession cults for the relationship between devotees and spirits to be expressed in the image either of marriage or of descent or sometimes of both. In the case of tumbura it is through the image of descent rather than through that of marriage that links are established in the ritual context between the devotees, past and present, although marriage

13 170 TUMBURA IN SUDAN symbolism is also present in the initiation ceremony.36 This is so because descent is of vital importance to the construction of a positive self-identity among the devotees. To understand this we should return to the description of the initiation ceremony. Before the slaughtering of the sacrificial sheep the sanjak performs the tahlil, a piece of instrumental music that stands for the expression 'There is no God but God'. During its performance the brigadar, the sanjak's male assistant, recites the tatriq, an oration that functions as the fatiha, the first chapter of the Quran. The tatrtq begins by mentioning the prophet Muhammad and Bilal, the first mu,adhdhin of Islam and 'father of all sanjaks', who, significantly, was of slave descent. It continues with Adam and Eve, Sufis, the past officiants and members of the cult, until it comes down to the novice. Then the brigadar slaughters the sheep and gives the novice some of its blood to drink. In these acts we have the symbolic introduction of the novice into the tumbura ritual descent system. Two points should be emphasised here. First, through the tatrfq the novice is linked with Adam and Eve and by extension with the human race. This is a theme that can be found among contemporary populations such as the Gumuz of the Sudan-Ethiopia border, who suffered considerably from slavery and appear in the context of tumbura.37 As they argue, 'We are sons of the eldest [sons of Adam], and that means we are freemen' (James, 1988: 135) and consequently humans. More than that, reminding us of the Sfdanif claim of being the 'original people', James maintains of the Gumuz that 'being regarded as sons of the firstborn... can be regarded as a prior claim to land...' (ibid.). Second, drinking the sacrificial blood transforms the novice into one of the awldd or bandt of tumbura, i.e. the sons or daughters of the spirit, and into a brother or sister of the other devotees. This image of family is completed by calling the shaykha ummiyya, i.e. 'mother',38 and the sanjak bdba, i.e. 'father'.39 Acquiring history: songs and spirit modalities The tumbura songs (jawab, literally 'reply' or 'message') performed by the sanjak during the ceremonies are the means through which the spirit is invited to present itself to the world. Under their compulsive potency the devotees get into a trance manifested through ecstatic dancing; but it is not they who dance, it is the spirit itself which uses their bodies as doors to enter the physical world. In this process of embodiment the spirit assumes a variety of forms (khayt, literally 'thread') which are acted out in a theatrical manner by its possessed hosts. These forms do not compromise the singleness of the spirit-as in bore the spirit is one and the khayts simply 'come from it'. It is for this reason I call these forms spirit modalities. The spirit modalities correspond to categories drawn from the historical context that makes the cult meaningful and represent the experience of the Sudanf community in history. As such, the modalities have been incorporated into tumbura in a gradual manner. This may sound identical with what I have said earlier about bore, but it is not. There the spirits are representatives of the Other, while in tumbura the modalities represent parts of the devotees' collective Self which is celebrated through possession. Denied by slavery, the humanity of the slaves is reconstituted in the space of the ritual.

14 TUMBURA IN SUDAN 171 The tumbura modalities include the Nuba, the Banda the Gumuz, the Sawakniyya, the Lambanat, the Bashawdt and the Khawdjat. The first three represent pagan tribes to which many devotees belonged before their enslavement. So greatly have these tribes suffered from the slave trade that their names have become synonyms for slave. The Sawdkniyya took their name from Sawakin, a nineteenth-century Red Sea port through which slaves were exported abroad. According to tradition, tumbura itself came from Sawakin, carried by the ex-slave soldiers of the Sudanese battalions who were stationed there in the Mahdist period.40 The Bdshawdt represent the Egyptian officers who served in Sawakin during the same period, while the Lambanat refer to female slaves from Sawakin. Lastly, the Khawdjat represent the British colonialists. The Khawajdt (and to an extent the Bdshawat) are perceived in an ambiguous manner: as infidel invaders similar to the Turco-Egyptians, yet as a force that brought freedom and dignity to the slaves. 'The English were good with the Sudanfnyya They came here and we had freedom. The Siudani could open shops and become traders like the Arabs.'41 Of the two, it is the view of the Khawdjdt as liberators that prevails. Surely the fact that many Sudanf cult devotees had served under British and Egyptian officers must have influenced this perception.42 The modalities which I have just described are intimately related, almost fused, with the songs which bring them forward during the ceremonies. These songs, like the modalities, must not be seen as 'a totality present in simultaneity' (Bourdieu, 1992: 38). They were composed gradually between the 1860s and the 1910s by sanjaks serving in the nineteenth-century slave armies of the Sudanese battalions, and represent a commentary on the life of the Sudannf, mentioning places, events and characters with special reference to slavery and battles.4 The last song was composed after the battle of Omdurman in Colonialism and the ban on slavery ushered in a new era and the end of further additions in tumbura in terms of modalities and songs. The songs, then, and their corresponding modalities are 'representations of pastness' embodying their own authenticity;44 they are parts of social action and cannot be regarded as historical data or documents outside their constitutive social relationships. Through them the Siudnf devotees see themselves as members of a distinct victimised community acting in history, and acquire a human voice to talk about this community which as mere slaves they did not have.45 As they say, the songs are the legacy that the sanjaks of the past, the 'Lords of the Old Truth' (Sayyid al-haqq al-qadfm), have bequeathed to them. For that reason, they claim, the songs are a chain (silsila, silik) that has always remained the same. 'There is no alteration [maff taghadil]; what exists exists [al-qa'da di bas ihd].' Although this denies the dynamism of tumbura and the fact that it has always been in the making, in viewing the modalities as aspects of the devotees' Sidani self, and their songs as expressions of their Sudandn history, the claim makes perfect sense: since there is one history only, there should be one way of talking about it through tumbura. For the Sudanf devotees, that and only that is important. The three processes discussed above show that tumbura is a mechanism of inclusion as well as of exclusion. On the one hand, it presents its subordinate

15 172 TUMBURA IN SUDAN devotees as members of humanity and devout Muslims just like their overlords through the creation of a ritual descent system that goes back to Adam and Eve and through the relation between tumbura and the Qadiriyya Sufi order. On the other hand, tumbura emphasises the difference between its people and the dominant Arabs. This is done through the songs of the cult, where the position of the Sudani in history as active agents rather than subhuman slaves is celebrated. The decisive moment in which this comes to the surface is during the performance of the last Khawdjat song, 'The Briton is the king of the Sudanf,' which the group members sing standing up while in their midst dances a devotee dressed like a British colonial officer. In this song what is really celebrated is the (mistakenly assumed) end of Arab violence, the end of slavery. Such being the case, tumbura should be seen as part of a Sudanf counterdiscourse that articulates an alternative reading of history and a positive selfidentity for its devotees. What is interesting, though, is that the way in which this is done reveals the existence of a contradiction within the tumbura discourse, the same contradiction which, as we have seen, distinguishes Sudanf self-identity: its structural predisposition towards its own undoing. By presenting its Sudanf devotees as similar to the Arabs, only to pronounce them different or even better a moment later, tumbura appropriates parts of the dominant tradition, in particular its Sufi orientation and its descent logic. In short, the tumbura people embrace part of the culture of those who deny them all culture in order to overcome that negation. Tumbura, then, marks similarities and retains differences; it presents devotees as members of humanity like the Arabs, and legitimates their status as 'original people' unlike the Arabs. It is a forum for Arabisation and a bastion of Sudanfness.46 THE DECLINE OF TUMBURA The tumbura discourse as presented above is constructed around the Arab-Sudanin distinction, a particular conceptualisation of the relations between dominant and subordinate populations that is rooted in the nineteenth-century slave trade. When the distinction ceases to be relevant to contemporary socio-political conditions, tumbura too becomes meaningless. It is on this proposition that I shall base my explanation of tumbura's steady decline during the present century. As has been argued, though the abolition of slavery led to the improvement of the position of the ex-slaves, it did not transform the negative attitudes of the traditionally slave-holding classes towards them, since culturally it did not effect their humanisation (Hargey, 1981: 318). Not only did the distinction between freemen and freed slaves remain very much alive, but it also widened its scope to include an ill-defined mass of non-arab Muslim migrants who over the years flooded the Northern cities.47 In these conditions, coupled with a new climate of colonial civil order after years of war and violence, tumbura flourished unprecedentedly, with more than eighty groups operating in the Greater Khartoum area.4 In reality the flowering was more apparent than real, as the number of initiations after 1898 was declining. This is illustrated by the fact that there

16 TUMBURA IN SUDAN 173 have been fewer and fewer people to replace dying cult officiants, as well as an increasingly higher attrition rate among those who have celebrated the thanksgiving ceremony, but have never been formally initiated into the cult. The reasons behind this trend are related to the process of defining Sudanese national identity. This process, although dominated by the Arab Northerners, was articulated in such a way as to attract-or, more correctly, to drag in-the African Southerners and other non-arab populations who constituted more than a third of the population of colonial Sudan.49 When the country became independent in 1956 the situation reverted to the old pattern, with class distinctions and social mobility predicated upon the principles of selfasserted Islamic pedigree and Arab descent on the part of the Northern Sudanese. This exacerbated existing economic disparities and the tremendous political, historical and cultural differences and led to the civil war between the North and the South to which I have already referred. The transformation of the non-muslim Southerners from enslavable populations beyond the administrative and moral periphery of nineteenthcentury Sudan into second-class Sudanese citizens, rebels and refugees (i.e. political actors operating from within the system) invalidated the Arab-Sidani distinction in two ways. First, it established the Southerners as a new category at the bottom of social hierarchy; second, it facilitated the gradual disintegration of the social category which subscribed to a Sidann self-identity. The bulk of the slave descendants and all those who identified themselves as Siudanf moved closer to the Arabs as Muslims vis-a-vis the non-muslim Southerners. This accentuated the existing Arabisation of the slave descendants, who found it easier than before to adopt Arab values or identities and to throw off their Siudanf-ness, forgetting that their ancestors were slaves.50 The whole process, which Doornbos (1988: 100) callsironically, from our point of view-'sudanisation',51 was also facilitated by demographichanges, namely the length of residence in the North, the death of those whose fathers or grandfathers had been slaves, and the participation of the younger generations in national educational programmes. Within this framework, tumbura membership did not offer something positive to the coveted upward mobility of slave descendants and other related groups. On the contrary, it indicated backwardness and association with 'less Sudanised' people who are still called slaves. Of the eighty or so groups in the 1950s and 1960s, today only five function in the Greater Khartoum area. Most of the cult members are elderly and are not being replaced by younger ones. As they argue: Those who know [tumbura] belong to our generation. Then it is the end. We are the last people [al-nas al-nihayya halds]. Our children say that tumbura is forbidden by Islam [haram]... that it is the work of crazy people [shugul majanin]. They prefer to go to the market wearing a clean jalabiyya [men's garment worn mostly by the Arabs]... they are clerks and belong to the Afendiyya [the educated middle and upper classes, by definition dominated by Northerners]. I cannot say whether the descendants of the older tumbura devotees are joining other cults, notably bore, the other Arab-dominated practice of the zdr cult complex, or whether they associate themselves with more 'orthodox'

17 174 TUMBURA IN SUDAN practices such as Sufism. It seems, however, that some of the remaining tumbura groups undergo what might be called 'boresation', becoming exclusively female and introducing practices which are present in boresigns, as I interpret them, of Arabisation. In certain cases it is only the presence of the sanjak playing the rababa and the singing of the cult's songs that signify the difference between the two cults. The young devotees of these 'boreised' groups fall into two categories. They are either children of migrants, mostly from the Nuba hills, who retain close ties with their villages and speak in their native idioms alongside Arabic; or, in a few cases, they are children of third or fourth-generation slave descendants who have almost forgotten their slave past. In both respects, to such people the Arab--Sudan distinction means little. They do not understand the songs of the cult, nor do they know its history. Many of them frequent simultaneously tumbura and bore, something which until recently was anathema to the older officiants who never held bore in respect. The fact also that the new groups are dominated by women may suggest that subordinate men of non-arab descent can be accommodated to the new environment more easily than women. This shows that gender becomes as important a parameter in tumbura as it is in bore, something that was not the case earlier on, when the emphasis was on non-gendered Suidanf self-identity. Boddy, then, is correct in arguing that zar-tumbura has now been 'subsumed' within zar-bori [sic] and that the two together are known simply as zdr (Boddy, 1989: 133). This, however, is a recent development which hides a world of difference whose measure has been the terror of slavery. CONCLUSION In the previous pages I have attempted to analyse the tumbura spirit possession cult as it was presented to me by its officiants and devotees in the area of Greater Khartoum in Central to the analysis were two conditions set by the nature of the cult itself. First, present within the cult are, and always have been, sub-discourses which construe it in interrelated yet distinct ways, each highlighting or obscuring specific aspects or facets of it. Analytically, in order to discuss tumbura as a process in time, I have distinguished between two such sub-discourses, and simultaneously I emphasised that my privileging one of them did not elevate it to the level of a perceived orthodoxy-although there was such a tendency among several of my interlocutors. The second condition arose from the fact that the work of Kenyon (1991a,b) in the region of Sinnar, and my own brief field experience among two cult groups in Port Sudan and Sawakin on the Red Sea,52 show that, beyond the existence of guiding principles that bind different cult groups together in a corporate organisation, tumbura groups operate within a relatively wide range of beliefs and practices-although I would suspect that this is a consequence of the cult's recent decline. To an extent, these conditions seem to relativise my analysis of tumbura and to limit its scope. On the other hand, they bring us closer to reality, a reality that is rooted in processes that far surpass the cult groups themselves or the shanty towns in which they can be found. These processes have to do with the construction of Sudanese national identity and, more specifically,

18 TUMBURA IN SUDAN 175 with the involvement in them of the slave descendants and other related subordinate groups, collectively called Sdanin. Tumbura was analysed as part of a Sdainf counter-discourse, i.e. as part of an answer the Sudanl people were forced to come up with in response to those who dominated the discussion on what it means to be Sudanese. As such, tumbura cannot be seen outside the concrete historical framework which encompasses the lives of its devotees, themselves 'agents of their times'. Rather, it is an inherent aspect of the 'lived in' political conditions of the everyday life of these people or, to paraphrase Kelly (1988: 53), the 'performative' accomplishment of the trajectory of their self-identity in real time and space. From such a point of view, tumbura can be analysed not only historically but in the present tense, because it is one field in which radical changes in the positioning of the Sudann people within the wider social formation are still taking place. History has no end, at least for the tumbura people,53 who say that tumbura's Old Truth will not be lost with the eventual death of its older devotees, al-nas al-nihdyya. It will always be there with the younger devotees in the 'boreised' groups, like the group headed by shaykka Nura, a middle-aged woman from the Nuba hills. 'When the last sanjak dies,' shaykha Nura told me, 'we will take the rababa [lyre] out on to the courtyard and we will sing the songs he taught us.' NOTES 1 See 2 Boddy (1994). For a critique see Wilson (1967), Bourguignon (1976), Crapanzano (1977), Giles (1987) and Boddy (1989). 3 For a critique see Gellner (1994: 29), Lambek (1985: 294, 300-2) and Lewis (1990: 59). 4 See, for example, Abu Lughod's (1993) critique of Boddy's (1989) work. 5 See, for example, Brown's (1991) analysis of Vodou lives in Haiti and Brooklyn; Caplan's (1975) study of possession on Mafia island; Gellner's (1994) article on Kathmandu possession cults and shamanism; Gow's (1994) analysis of shamanism in Western Amazonia; Kramer's (1993) discussion of possession in relation to the conceptualisation of the Other and its representation in African art; Lewis's (1986) discussion of African 'survivals' in Islam; Lewis et al., (1991) articles on the zar-bori cult; and Stoller's (1989) account of possession among the Songhay of Niger. 6 For the Arabisation of the Northern Sudan see Yisuf Fadl Hasan (1967). 7 I borrow the term from Anderson (1992). 8 See also Bj0rkelo (1989: 55). The slaves are Kramer's (1993) barbarians to whom were '... denied precisely those general attributes which are felt to be intrinsically and purely human' (p. 6). 9 See also Ahmad al-shahi (1972: ). A more general discussion concerning the ideology of Islamic slavery can be found in Willis 10 (1985: esp. 11). See also Levtzion (1985: 189) and Lewis (1966, 1986: 97). These trends did not cease with the end of the slave era but continued well into the twentieth century. See Holy (1991); James (1971); Manger (1994). l Cf. Bloch (1980) and Twaddle (1993) for similar cases elsewhere. 12 Although the term 'Arab' cannot be always defined with the necessary precision, it does nonetheless convey a valid political appraisal. 13 See Daly (1986). 14 The Sudanese were also called volunteer slaves (!), servants or unpaid workers (Hargey, 1981: 75). See also Hargey (1981: 75, 102, , 475-6); Makris (1994b, 1995) and McLoughlin 15 (1970: 105). For a recent contribution on this see Salah El-Zein (1995).

19 176 TUMBURA IN SUDAN 16 For the British 'Southern Policy' which kept the South separated from the North until the closing days of the colonial period see Daly (1991). 17 Strictly speaking, this started in 1947 with the Juba conference that was concerned with the 'mechanics... [of] the south's future in union with the north' (ibid.: 240). 18 A recent violent fragmentation of the Southerners' SPLA movement along tribal lines shows that this 'monolithic' nature of the South is largely imaginary; it can be entertained, if at all, only in contradistinguishing the Southern peoples from the Northerners. 19 The tribes mentioned were on the frontiers of the nineteenth-century slave trade. 20 It is interesting that the terms used, 'Arab' and 'Siidani', are identical to those employed by the British colonialists. 21 Cf. Deng (1987: 61, 65); Fouad N. Ibrahim and Rupert (1988). 22 Cf. Ewald (1990: 137-8), Holy (1974: 54; 1991: 19 ff.), James (1971), Kurita (1995: 13), O'Fahey (1979: 202) and Tully (1988). 23 For zar in other Muslim countriesee Kennedy (1967), Lewis (1989), Lewis et al. (1991), Morsy (1978, 1991) and Saunders (1977). 24 Seligman herself did not realise that what she saw was tumbura. 25 Seligman's (1914) descriptions of tumbura come from the quarters of the Sudanese battalions. For the role of soldiers in tumbura see Kenyon (1991a,b) and Makris (1994b). 26 See Kenyon (1991a,b). 27 Pains, haemorrhages, those more often cited. 28 See 29 Kenyon (1991a: 199). Bourdieu paralysis, blindness and psychological disturbances are among (1992: 3843); Comaroff & Comaroff (1992: 53). 30 The two political parties which have dominated Sudanese politics since independence almost correspond with two religious brotherhoods. 31 With the possible exception of the Qadiriyya and the Khatmiyya. 32 A similar assertion for Bore can be found in Constantinides (1977: 64) and Samia al-hadi al-nagar (1980: 686). For the similar 'Black Brotherhoods' of northern Africa see Ferchiou (1991) and Lewis (1989: 93). 33 The relation between these two orders and tumbura may be explained by the fact that the Qadiriyya has always been open to African influences more than other orders, and the Khatmiyya has been particularly active among ex-slave-soldiers. 34 Ahmad al-rifa'f and Ahmad al-tijani are the founders of the Rifa'iyya and the Tijaniyya brotherhoods. Muhammad al-hasan b. Muhammad 'Uthman al-mfrghanf (d. 1869) was the person 'who successfully consolidated the Khatmiyya..' brotherhood (Ali Salih Karrar, 1992: 75). His grandson, Sayyid Alf al-mfrghanf (d. 1968), a celebrated religious figure, played an important role in Sudanese politics. For the Khatmiyya rituals see Ali Salih Karrar (1992: ) and Trimingham (1983: ). 36 See Bloch (1994: 136) for the creation of ritual descent among possessed slaves in Madagasar. 37 See below. 38 See Kenyon (1991a: 185). 39 See Brown (1991: 48). 40 All the sanjaks I met, as well as their fathers and grandfathers, were in the Sudanese battalions. 41 Cf. Abbas Ibriahm Muhammad Al (1972), Hargey (1981) and McLoughlin (1962). 42 See Hill and Hogg (1995: ix). 43 The similarities between the tumbura songs and the so-called Sudanese soldiers' songs as they appear in Thorburn (1925) are indicative of the connection between the cult and the army. See Makris 44 (1994a). See Tonkin (1992: 8, 50-65). 45 Cf. Barber (1989). See also Friedman (1992: 195). 46 Cf. de Moraes Farias (1989). 47 The cleavage between Arab and non-arab or 'black' Muslims in the Sudan can be seen in the wider context of Muslim politics in sub-saharan Africa, where, as Brenner (1993: 20) argues, 'despite the insistence by the ideologues of the universalistic umma that Islam dissolves all differences of race, black Muslims in Africa are finding unity in Islam as blacks'.

20 TUMBURA IN SUDAN Ahmed el-safi (1988) gives a list of sixty-nine tumbura leaders who headed groups located in the area of Greater Khartoum; in 1988 eleven of them were still alive. Discussing this list with several cult devotees in , we were able to add another thirteen names, coming to eighty-two. There was general agreementhat most of the groups were active during the colonial period and the first decade after independence in See El-Fatih A. Abdel Salam (1989: 47-8); Mahasin Abdelgadir Hag al-safi (1989); Muddathir Abd al-rahfm (1986: 89 ff.). Again the distinction between North and South is of a cultural and political nature. 'Indeed, there were certain areas in the North... which were as backward as certain areas in the South. These areas, however, formed part of a more homogenou society in the North by virtue of their religion, culture and historical experience' (Mohamed Omer Bashir, 1968: 70). 50 As Bloch (1992: 13) argues, 'When we consider the social actor's attitude to the distant past it becomes clear that one's effort involves not simply finding ways of remembering better... but also, and equally, finding ways to forget it.' 51 Since the 1920s the term Sudani has gradually lost its earlier meaning and come to signify the Sudanese national whose cultural outlook is that of the Arab Northerners. 52 These two groups were even more 'traditional' than those in the Greater Khartoum area, if I may put it that way. 53 Cf. Bawa Yamba (1995: 32 ff.). REFERENCES Abbas Ibrahfm Muhammad Alf The British, the Slave Trade, and Slavery in the Sudan, Khartoum: Khartoum University Press. Abu Lughod, L Review of J. Boddy, Wombs and Alien Spirits, American Ethnologist, 29 (2), Ahmed al-safi 'The Tumbura Cult'. Unpublished paper presented at the Zar Workship, Institute of African and Asian Studies, Khartoum. Ahmed al-shahi 'Spirit possession and healing', Bulletin, British Society for Middle Eastern Studies, 11 (1). Ali Salih Karrar The Sufi Brotherhoods in the Sudan. London: Hurst. Anderson, B Imagined Communities. London: Verso. Ardener, E 'Belief and the problem of women', in J. S. La Fontaine (ed.), The Interpretation of Ritual. London: Tavistock. Barber, K 'Interpreting Orfki as history and as literature', in K. Barber and P. Farias de Moraes (eds.), Discourse and its Disguises: the interpretation of African oral texts, pp African Studies Series 1, Birmingham: University of Birmingham. Bjorkelo, A Prelude to the Mahdiyya: peasants and traders in the Shendi region, Cambridge: Cambridge University Press. Bloch, M 'Modes of production and slavery in Madagascar: two case studies', in J. Watson (ed.), Asian and African Studies on Slavery, pp Oxford: Blackwell 'Internal and external memory: different ways of being in history', Suomen Antropologi 1, 'The slaves, the king, and Mary in the slums of Antananarivo', in N. Thomas and C. Humphrey (eds.), Shamanism, History and the State, pp Ann Arbor: University of Michigan Press. Boddy, J Wombs and Alien Spirits. Madison: University of Wisconsin Press 'Spirit possession revisited: beyond instrumentality', Annual Review of Anthropology 23, Bourdieu, P Outline of a Theory of Practice. Cambridge: Cambridge University Press. Bourguignon, E Possession. San Francisco: Chandler.

21 178 TUMBURA IN SUDAN Brenner, L. (ed.) Muslim Identity and Social Change in sub-saharan Africa. London: Hurst. Brown, K Mama Lola: a Vodou priestess in Brooklyn. Berkeley: University of California Press. Caplan, A. P Choice and Constraint in a Swahili Community. London: Oxford University Press for the International African Institute. Comaroff, J. L., and Comaroff, J Ethnography and the Historical Imagination. Boulder: Westview Press. Constantinides, P 'Sickness and the Spirits: a study of the Zar spirit possession cult in the northern Sudan'. Unpublished Ph.D. thesis, University of London 'Ill at ease and sick at heart: symbolic behaviour in a Sudanese healing cult', in I. M. Lewis (ed.), Symbols and Sentiment, pp London: Academic Press 'Women's spirit possession and urban adaptation in the Muslim northern Sudan', in A. P. Caplan and J. Bujra (eds.), Women United Women Divided, pp Bloomington: Indiana University Press 'The history of Zar in the Sudan: theories of origin, recorded observation and oral tradition', in I. M. Lewis, Ahmed al-safi and Sayyid Hurreiz (eds.), Women's Medicine: the zar-bori cult in Africa and beyond, pp Edinburgh: Edinburgh University Press, for the International African Institute. Cooper, F 'Islam and cultural hegemony: the ideology of slave owners on the East African coast', in P. E. Lovejoy (ed.), The Ideology of Slavery in Africa, pp London: Sage. Crapanzano, V The Hamadsha. Berkeley: University of California Press 'Introduction', in V. Crapanzano and V. Garrison (eds.), Case Studies in Spirit Possession, pp New York: Wiley Tuhami: portrait of a Moroccan. Chicago: University of Chicago Press. Daly, M Empire on the Nile. Cambridge: Cambridge University Press Imperial Sudan. Cambridge: Cambridge University Press. Deng, F. M 'Myth and reality in Sudanese society', in F. M. Deng and P. Gifford (eds.), The Search for Peace and Unity in the Sudan. Washington D.C.: Wilson Center Press. Doornbos, P 'Becoming Sudanese,' in T. Barnett and Abbas Abd al-karim (eds.), Sudan: state, capital, and transformation. New York: Croom Helm. El-Fatih, A. Abdel Salam 'Ethnic politics in the Sudan', in Sayyid H. Hurreiz and El-Fatih A. Abdel Salam (eds.), Ethnicity, Conflict and National Integration in the Sudan, pp Khartoum: Institute of African and Asian Studies. Ewald, J. J Soldiers, Traders, and Slaves: state formation and economic transformation in the greater Nile valley, Madison: University of Wisconsin Press. Farias, P. F. de M 'Pilgrimages to "pagan" Mecca in Mandinka stories of origin reported from Mali and Guinea-Conakry', in K. Barber and P. F. de M. Farias (eds.), Discourse and its Disguises: the interpretation of African oral texts, pp Birmingham: Birmingham University African Studies Series 1. Ferchiou, S 'The possession cults of Tunisia: a religious system functioning as a system of reference and a social field for performing actions', in I. M. Lewis, Ahmed al-safi and Sayyid Hurreiz (eds.), Women's Medicine: the zar-bori cult in Africa and beyond, pp Edinburgh: Edinburgh University Press, for the International African Institute. Fouad N. Ibrahim and Rupert, H. (eds.) Rural-Urban Migration and Identity Change: case studies from the Sudan. Bayreuth: Druckhaus Bayreuth. Friedman, J 'Myth, history and political identity', Cultural Anthropology 7 (2),

22 TUMBURA IN SUDAN 179 Gellner, D. N 'Priests, healers, mediums and witches: the context of possession in the Kathmandu valley, Nepal', Man (n.s.) 29 (1), Giles, L. L 'Possession cults on the Swahili coast: a re-examination of theories of marginality', Africa 57 (2), Glassman, J 'The bondsman's new clothes: the contradictory consciousness of slave resistance on the Swahili coast', Journal of African History 32. Gow, P 'River people: shamanism and history in western Amazonia', in N. Thomas and C. Humphrey (eds.), Shamanism, History and the State, pp Ann Arbor: University of Michigan Press. Hargey, T 'The Suppression of Slavery in the Sudan, '. Unpublished Ph.D. thesis, St Antony's College, Oxford. Hayder Ibrahim The Shaiqiya: the cultural and social change of a northern riverain people. Wiesbaden: Franz Steiner. Hill, R Egypt in the Sudan, Oxford: Oxford University Press. Hill, R., and Hogg, P A Black Corps d'elite: an Egyptian Sudanese conscript battalion with the French army in Mexico, , and its survivors in subsequent African history. East Lansing: Michigan State University Press. Holt, P. M The Mahdist State in the Sudan. Oxford: Oxford University Press. Holt, P. M., and Daly, M. W A History of the Sudan. London: Longman. Holy, L Neighbours and Kinsmen. London: Hurst Religion and Custom in a Muslim Society. Cambridge Studies in Social and Cultural Anthropology, Cambridge: Cambridge University Press. Jackson, H. C Behind the Modern Sudan. London: Macmillan. James, W 'Social assimilation and changing identity in the southern Funj', in Yiisuf Fadl Hasan (ed.), Sudan in Africa, pp Khartoum: Khartoum University Press 'Perceptions from an African slaving frontier', in L. Archer (ed.), Slavery and other Forms of Unfree Labour, pp London: Routledge. Kapferer, B A Celebration of Demons: exorcism and the aesthetics of healing in Sri Lanka. Bloomington: Indiana University Press. Kehoe, A., and Giletti, D 'Women's preponderance in possession cults: the calcium deficiency hypothesis extended', American Anthropologist 83 (3), Kelly, J. D 'From Holi to Diwali in Fiji: an essay on ritual and history', Man (n.s.) 23, Kennedy, J. G 'Nubian Zar ceremonies as psychotherapy', Human Organisation 26 (4), Kenyon, S. 1991a. Five Women of Sennar; Oxford: Clarendon Press. 1991b. 'The story of a tin box: Zar in the Sudanese town of Sennar', in I. M. Lewis, Ahmed al-safi and Sayyid Hurreiz (eds.),'women's Medicine: the zar-bori cult in Africa and beyond, pp Edinburgh: Edinburgh University Press, for the International African Institute. Kramer, F The Red Fez: art and spirit possession in Africa. London: Verso. Kriss, R., and Kriss-Heinrich, H. 1960, Volksglaube im Bereich des Islam. Wiesbaden: Otto Harrassowitz. Kurita, Y. (1995) 'The Language of Class and the Language of Race in the moder Sudanese politics: the case of Alf Abd al-latff and the Revolution of 1924'. Unpublished paper presented at the Symposium on Cultural Diversity and Nation Building in the Sudan, University of Cairo. Lambek, M Human Spirits: a cultural account of trance in Mayotte. Cambridge: Cambridge University Press 'Ecstasy and agony in Sri Lanka: a review article', Comparative Studies in Society and History 27 (2),

23 180 TUMBURA IN SUDAN Levtzion, N 'Slavery and Islamisation in Africa: a comparative study', in J. R. Willis (ed.), Slaves and Slavery in Muslim Africa I, pp London: Cass. Lewis, I. M 'Spirit possession and deprivation cults', Man 1, Ecstatic Religion. Second edition 1989, London: Routledge. (ed.) Symbols and Sentiments: cross-cultural studies in symbolism. London: Academic Press Religion in Context: cults and charisma. Cambridge: Cambridge University Press 'Spirits at the house of childbirth', Times Literary Supplement, 1-7 June, p Lewis, I. M., Ahmed al-safi and Sayyid Hurreiz (eds.) Women's Medicine: the zar-bori cult in Africa and beyond. Edinburgh: Edinburgh University Press, for the International African Institute. Mahasin Abdelgadir Hag al-safi The Nationalist Movement in the Sudan. Khartoum: Institute of African and Asian Studies. Makris, G. 1994a. 'Creating history: a case from the Sudan', Sudanic Africa 5, b. 'Creating Ethnicity: the case of the Sudanl people.' Unpublished paper presented at the conference on Ideologies of Race, Origins and Descent in the History of the Nile Valley and North East Africa, St Antony's College, Oxford 'The Construction of Categories: from the era of colonialism to the days of militant Islam'. Unpublished paper presented at the symposium on Cultural Diversity and Nation-building in the Sudan, University of Cairo. Manger, L From the Mountains to the Plains: the integration of the Lafofa Nuba into Sudanese society. Uppsala: Nordiska Afrikainstitutet. McLoughlin, P. F. M 'Economic development and the heritage of slavery in the Sudan republic', Africa 32, 'Labor market conditions and wages in the Three Towns, ', Sudan Notes & Records LI, Messing, S 'Group therapy and social status in the Zar cult of Ethiopia', American Anthropologist 60 (6), Mohamed Omer Bashir The Southern Sudan: background to conflict. London: Hurst. Morsy, S 'Sex roles, power and illness in an Egyptian village', American Ethnologist 5, 'Spirit possession in Egyptian ethnomedicine', in I. M. Lewis, Ahmed al- Safi and Sayyid Hurreiz (eds.), Women's Medicine: the zar-bori cult in Africa and beyond, pp Edinburgh: Edinburgh University Press for the International African Institute. Muddathir Abd al-rahfm Imperialism and Nationalism in the Sudan. Khartoum: Khartoum University Press. Obeyesekere, G Medusa's Hair: an essay on personal symbols and religious experience. Chicago: University of Chicago Press. O'Fahey, R. S 'Islam, state and society in Dar Fur', in N. Levtzion (ed.), Conversion to Islam, pp New York: Holmes & Meier 'Fur and Fartft: the history of a frontier', in J. Mack and P. Robertshaw (eds.), Culture History in the Southern Sudan, pp Memoir No. 8 of the British Institute in Eastern Africa, London: Thames & Hudson 'The question of slavery in the Sudan', in Slaveri af avvikling i komparativt perspectiv, pp Trondheim. Rasmussen, S. J Spirit Possession and Personhood among the Kel Ewey Tuareg. Cambridge: Cambridge University Press.

24 TUMBURA IN SUDAN 181 Salah El-Zein 'Articulation of Cultural Discourse and Political Dominance in Sudan', Sudanese Alternative Discourses (SAD) 2, fatihservtech.com. Samia al-hadi al-nagar 'Spirit Possession and Social Change in Omdurman'. Unpublished M.Sc. thesis, University of Khartoum 'Zaar practitioners and their assistants and followers in Omdurman', in V. Pons (ed.), Urbanization and Urban Life in the Sudan, pp Khartoum: Development Studies and Research Centre, University of Khartoum, and Hull: Department of Sociology and Social Anthropology, University of Hull 'Women and spirit possession in Omdurman', in S. Kenyon (ed.), The Sudanese Woman, pp Graduate College Publications, No. 19, University of Khartoum, London: Ithaca Press. Saunders, L. W 'Variants in Zar experience in an Egyptian village', in V. Crapanzano and V. Garrison (eds.), Case Studies in Spirit Possession. New York: Wiley. Scott, J. C Domination and the Arts of Resistance. New Haven: Yale University Press. Seligman, B. Z 'On the origins of the Egyptian Zar', Folklore 25, Seligman Collection of the Royal Anthropological Institute, Box II, A. 4. Stiansen, E 'Overture to Imperialism'. Unpublished Ph.D. thesis, University of Bergen. Stoller, P Fusion of the Worlds: an ethnography of possession among the Songhay of Niger. Chicago: University of Chicago Press. Thorburn, D. H 'Sudanese soldiers' songs', Journal of the African Society 23, Tonkin, E Narrating our Pasts. Cambridge: Cambridge University Press. Trimingham, J. S Islam in the Sudan. Oxford: Oxford University Press. Tully, D Culture and Context in the Sudan: the process of market incorporation in Dar Masalit. New York: State University of New York Press. Twaddle, M. (ed.) The Wages of Slavery. a Special issue of the journal Slavery and Abolition. Warburg, G. R 'Slavery and labour in the Anglo-Egyptian Sudan', Asian and African Studies 12, Historical Discord in the Nile Valley. London: Hurst. Ward, C 'Possession and exorcism: psychopathology and psychotherapy in a magico-religious context', in C. Ward, Altered States of Consciousness and Mental Health: a cross-cultural perspective, pp Newbury Park: Sage. Willis, J. R 'The ideology of enslavement in Islam', in J. R. Willis (ed.), Slaves and Slavery in Muslim Africa I, pp London: Cass. Wilson, P. J 'Status ambiguity and spirit possession', Man (n.s.) 2 (3), Yamba, C. Bawa Permanent Pilgrims: the role of pilgrimage in the lives of West African Muslims in Sudan. Edinburgh: Edinburgh University Press, for the International African Institute, and Washington, DC.: Smithsonian Institution Press. Yusuf Fadl Hasan The Arabs and the Sudan from the Seventh to the Early Sixteenth Century. Edinburgh: Edinburgh University Press. Zenkovsky, S 'Zar and Tambur as practised by the women of Omdurman', Sudan Notes and Records 31 (1), ABSTRACT The aim of this article is to study the tumbura spirit possession cult of the Sudan as a historical phenomenon directly associated with the social and political changes of the

25 182 TUMBURA IN SUDAN last hundred years. Tumbura can be found in the poor neighbourhoods and surrounding shanty towns of the big urban centres of northern Sudan. The majority of the cult's devotees are descendants of nineteenth-century African slaves who had been brought as slaves to the north from the southern and western Sudan by Arab Muslim northerners. Their conversion to Islam notwithstanding, the slaves and their descendants have always been regarded by the Arab northerners as subhumans with no religion, descent or history. In tumbura these claims are contested. Initiation into the cult enables the devotees to construct a positive self-identity and assert that they too are Muslims, descendants of Adam and Eve, and a social category with a specific sense of history-in other words, human beings. During the colonial and postcolonial periods the position of the slave descendants within the social formation changed considerably as new groups of people made an appearance. These were the refugees from the south who had fled famine and civil war and today populate the shanty towns that surround the capital, Khartoum, and other cities in the north. The refugees enabled the slave descendants to move away from the margins of the society and closer to their erstwhile Arab masters. This movement has triggered a process of identity transformation which has reflected on the popularity of tumbura, which has started to decline rapidly. At the same time the remaining groups are in a process of structural reorganisation which follows the lines of zdr bore, another, more popular, female cult practised mainly by people of Arab descent. RESUME Le but de cet essai est d'etudier le culte Tumbura de la possession de l'esprit au Soudan en tant que phenomene directement associe avec les changements sociaux et politiques des cent dernieres annees. On peut trouver le Tumbura dans les voisinages pauvres et les bidonvilles qui entourent les larges centres urbains du Nord du Soudan. Le majorite des fideles du culte sont des descendants d'esclaves africains du sud et de l'ouest du Soudan qui avaient ete amene comme esclaves au nord par les arabes mulsumans du nord. Malgre leur conversion a l'islam, les esclaves et leurs descendants ont toujours ete regarde par les arabes du Nord comme etant soushumains, sans religion, sans descendance ou histoire. Ces revendications sont contestees par le Tumbura. L'initiation au culte permet aux fideles de construire un sens d'identite positive et de faire valoir qu'eux aussi sont mulsumans, descendants d'adam et Eve, et qu'ils sont une categorie sociale avec un sens d'histoire specifique, en d'autres termes, qu'ils sont des etres humains. Pendant la periode coloniale et post-coloniale, la situation des descendants des esclaves au sein de la formation sociale a change considerablement quand de nouveaux groupes de gens sont apparu. Ceux-ci etaient les refugies du sud qui avait fuit la famine et la guerre civile et qui aujourd'hui peuplent les bidonvilles entourant la capitale, Khartoum, et les autres villes du nord. Les refugies ont permit aux descendants des esclaves de s'eloigner des marges de la societe et d'etre plus pres de leurs maitres arabes tout puissant. La transformation de l'identite qui a suivit a ete refletee dans la rapide diminution de la popularite du Tumbura. En meme temps, les groupes qui reste au Tumbura sont en train de connaitre un processus structurel de reorganisation qui suit celui de Zar Bore, un autre culte f6minin, plus populaire, qui est surtout pratique par les gens de descendance arabique.