POPULAR AND FORMAL ISLAM, AND SUPRA-LOCAL RELATIONS
The highlands of northwestern Tunisia, 1800-1970
Wim van Binsbergen
In recent years anthropologists have paid considerable attention to North African popular religion and to local saints as one of its main aspects. North African religious studies are in a transitory state in that anthropologists only recently have come to explore a field of enquiry which hitherto had been mainly worked by historians and Arabists. Coupled to the fact that, with such pioneers as Westermarck, Montagne, Evans-Pritchard, Peters and Bergue, the anthropology of North Africa made a comparatively late start in general (as compared with e.g. sub-Saharan Africa, or Oceania), this goes some way to explain why anthropological studies of North African region have tried to do rather too much. Instead of presenting a thorough descriptive analysis of specifically religious institutions, they tend to formulate highly abstract interpretative models which attempt to bring to bear an only cursorily-described religious system upon such topics as the variability and versatility of Islam , theories of segmentation, the state, society as a whole, folk illness, or the old controversy of the utilitarian versus the logical nature of religious symbols. Perhaps this state of affairs also reflects the allegedly world-wide tendency that anthropologists studying religion have been more interested in religious models than in religious behaviour.
In particular, some of the principal aspects of North African rural religion have hardly been discussed in anthropological writing so far: the worship of that cAtegory of land shrines that are not or only dimly associated with personal, historical saints; ecstatic cults of affliction, and in general the actual functioning of religion, as religion, within the social process at the village level.
However, various researchers did carry out fieldwork on these subjects and before long publications can be expected which fill in this gap. Meanwhile, it seems opportune to pursue, in the present paper, what is perhaps the most significant and mature line of enquiry within the already available studies: the relation between two major versions of Islam, one formal, the other popular, which have prevailed in North Africa over the centuries. It is on this subject that the recent Moroccan studies have made their most valuable contributions.
It is, then, the aim of this article to discuss, with a view on supra-local relations and incorporation processes, religious structure and change elsewhere in North Africa: in the highlands of Khumiriya, North-Western Tunisia.
Apart from small Jewish and Christian minorities, the whole of North Africa is nominally Islamic. Dominating in city life is the Quran. It imposes obligatory prayer (both at home and at the mosque), fast, alms, and the pilgrimage to Mecca, and is, moreover, associated with food prohibitions, certain general festivals, and an elaborate system of theology (commemorating the Prophets life-history) and law. In the rural communities; however, a popular, less formal and less strict version of Islam dominates; emphasizing saint worship (with great saintly festivals featuring in the agricultural calendar and eclipsing the general Islamic ones), ecstatic cults centring on affliction, and religious brotherhoods. These popular aspects are by no means absent in the cities; but whereas in the cities they exist only in the shadow of, and are incessantly challenged by, the urban formal version which makes a claim of constituting orthodoxy, in the rural areas the popular version makes up the local religion par excellence.
The outlines of the religious history of North Africa are well-known Ever since the Arab conquest in the seventh century, a recurrent theme has been the attempt to effectively spread the formal version of Islam from the urban centres into the rural areas.
This paper explores the interplay between local popular Islam and the repeated introduction of formal Islam in Khumiriya, against the background of its social and political structure and the radical changes the latter underwent in the colonial and post-colonial era. Having had no access to archival material on the area, my data derive from three sources mainly: participant observation, a systematic survey of present-day religious activities, and oral-historical research going back to about 1800.
My argument will suggest that, even if more detailed descriptive local information on North African rural religion is badly needed, the dynamics underlying the relation between the two versions of Islam in this part of the world should be interpreted primarily by reference to supra-local political and economic incorporation processes, i.e. to ultimately non-religious factor. This position clearly owes much to Gellners work - even though in the past I have criticized Gellner for reducing the specifically religious aspects of the North African saints to a marginal phenomenon.
2. The Structure of pre-colonial Khumiri Society in the Nineteenth Century
In pre-colonial, nineteenth century Khumiriya the economic and political basic unit was the homestead, a cluster of tents, which usually had the following composition: a middle-aged man; one or more married sons; one or more sons-in-law; and the wives and children of these men. Often the homestead also included one or a few unmarried clients: stranger herdsmen who in many cAses were to marry their patrons daughters. The homesteads were thinly scattered over the land. The density of population was about 12 per km2, less than 20% of the present one. Limited pressure on the land made possible a fluid pattern of semi-annual transhumance and short-distance migration geared to a near-subsistence economy revolving around animal husbandry and the cultivation of food-crops on forest clearings.
In this society honour and individual independence were central values, which cAme to the fore in great and often violent conflicts both within the homestead and between homesteads. Homesteads were involved in competition over women, animals, honour, and sometimes land. the homestead was far from stable. In the course of years it would dissolve: part of its membership remaining in the same territory while the other part would migrate to elsewhere (usually within a radius of 10 km).
The differences in wealth, authority and honour between the heads of the homesteads were limited, and fluctuating. Pre-colonial Khumiriya belonged to that type of society where an acephalous, segmentary social organization tends to develop. Accordingly, the limited data on the recruitment of partners in economic co-operation, conflict and religious activities in these old days suggest a segmentary pattern: social mobilization followed a tree-like structure of units at a series of levels, with units at one level being mutually exclusive but all nested within wider units at a higher level.
The classic anthropological segmentation model, which such authors as Favret and Gellner have applied to North Africa, hinged on unilineal descent. More in line with Peters penetrating criticism of this approach., segmentation in Khumiriya was (and is) more a matter of geographical propinquity than of unilineal descent. the homesteads were part of wider territorial segments (hamlets, villages, values, tribes), each associated with a particular, contiguous part of the land. Threshing-floors, springs, mens assembly-grounds (raquba), shrines and cemeteries functioned as visible attributes of distinct segments on each segmentary level. Each homestead would have its own threshing-floor but would combine with others in the use of a spring; the hamlet thus formed would combine with other adjacent hamlets in the use of the same mans assembly-ground; and on yet higher segmentary levels (valley, clan, tribe), villages thus formed would combine in their use of the same shrine and cemetery. the visible attributes indicative of a units segmentary level would be distributed and redistributed in accordance with the numerical size and power of the segments involved, and alterations therein. A group might expand from homestead level (having only its own threshing-floor to boast) to hamlet or village level (monopolizing a local spring and creating its own assembly ground) - or dwindle along the same scale. Threshing-floors, springs and mens assembly-grounds were however more than helpful markers in the anthropologists segmentary tree-diagram. Their serving as such is based on the fact that they were, pragmatically, the foci and vital economic and social processes: food production and the water supply focused on threshing-floor and spring, whereas the mens assembly-ground was the major arena for the on-going social and local political process. Apparently, the structure of segmentation directly sprang from the Dynamics of daily life. However, on the highest segmentary levels of valley, clan, and tribe, segmentary attributes were used (shrine, cemetery) that had no pragmatic function in everyday life.
Although the homesteads had by no means a purely agnatic composition, and although the basic pattern of social organization was based on locality rather than descent, yet a powerful agnatic ideology existed which still provides a dominant cultural idiom in Khumiri society. this ideology implies that effective positive relationships should be ideally formulated as relationships between (close)agnates. If interacting people are actually not agnatic Ally related, or not related at all, fictive agnatic ties have to be created through genealogical manipulation.
Therefore, persons who had been living in one anothers proximity for some decades, would be affiliated to the same mythical ancestor (apical ancestor of a clan named after him) irrespective of objective, historical, genealogical links. On the other hand persons who shared, historically, the same matrilineal ancestors would cease to be considered close agnates and would even no longer be reckoned to the same clan, if because of migration following the fission of homesteads they had not been living in one anothers proximity for several decades.
Thus if genealogical manipulation could ever be carried to the end, the result would e that clans and territorial segments would coincide. Proximity, economic and political co-operation, and intra-local marriages, would provide local integration of interaction, and thus in turn would be supported by the notion of common matrilineal descent. However, because of continuous migration, genealogical manipulation was always in a state of flux. Very recent immigrants would not yet be fully integrated in the locally dominant clan: hence temporarily certain brother segments, as accepted to belong to this dominant clan in their territory of origin; thus clan-affiliation temporarily provided identification between homesteads that were several kilometres apart: a condition cutting across territorial segmentation.
By having a powerful homestead, by establishing dyadic exchange relations with members of other homesteads, and by co-ordinating activities (fighting, conflict settlement, marriage negotiations) involving a wide social field, some heads of homesteads, and by co-ordinating activities (fighting, conflict settlement, marriage negotiations) involving a wide social fields, some heads of homesteads built up a position of great authority as elder (kabir, shaykh). Elders formed councils on several segmentary levels. Conflict regulation was their main task. However, in the most important, violent conflicts they would often belong to one of the parties, and then one had to resort to religious specialists (vide infra). In addition there existed, formally, an administrative structure of a higher order. In pre-colonial nineteenth-century Tunisia, each tribe had a qaid, appointed by the Bey of Tunis, and in charge of jurisdiction and taxes. In these days, Khumiriya nominally had its qaids as well, but their power in conflict regulation was extremely limited and Khumiris violently resisted paying taxes.
3. Aspects of popular religion in the pre-colonial era
Pre-colonial Khumiri society showed a continuous oscillation between territorial integration, and migration. As attributes of segments, shrines (and the cemeteries which surrounded the major shrines) played a very important role in this process.
Immigrants could settle in basically two ways. Either they arrived as clients of dominant, earlier inhabitants; in that case they had to orientate themselves (for sacrifices, pilgrimages, burials, oaths, and festivals) towards the shrines and cemeteries of their new patrons. Or they settled independent of earlier groups, on a territory which because of purchase, exchange, gift or violent conquest, on the moment of immigration was not occupied by a dominant other group. Then the immigrants would create a new shrine, often as a branch of the shrine of their original segment; - in the latter case the new shrine would be erected upon relics brought from there, and be given the same name. Initially, immigrants of either type would keep visiting the great saintly festivals of their segments of origin but before long these historical links would lapse; then immigrants and locals would jointly orientate themselves almost exclusively towards their (old or new) local shrines. the regular patter of territorial segmentation was restored - until further migrations demanded a new adaptive redistribution of shrines over segments.
In this way shrines formed the major visible beacons in the process of migration and territorial segmentation; the more so, as they were the only permanent buildings amidst the movable tents. At the same time they formed the beacons in the competition between segments. the invisible saint who, through his shrine, was associated with a certain segment, was supposed to give this segment his exclusive, and mighty help. Mass saintly festivals (where hundreds of visitors brought the ingredients for a collective meal, to be prepared, distributed and eaten in front of the shrine) provided a segment with an opportunity to show its wealth, strength, and allies. The visitors held a safe-conduct backed up by supernatural sanctions; therefore the festivals could be the only mass activity in this society otherwise so dominated by violence and divisiveness. Yet old disputes between segments tended to revive precisely on festivals. Because all higher-level segments endeavoured to embellish their shrine, to heighten the splendour of its festival, and to have that shrine accepted by more and more neighbouring segments as their focus of common ritual (thus making that shrine the attribute of an ever wider segment, on an ever higher segmentary level), the ecological competition between segments was, in many ways, duplicated by competition between local shrines. The history of shrines is, to a great extent, the history of their segments.
There was a close connexion between the worship of local shrines and the marriage pattern. Saintly festivals were marriage-markets. Moreover, women who had married outside their own village (as many as fifty per cent, or more, of all married women), were obliged to regularly visit the shrine of their village of origin. This norm was enforced by serious sanctions, both supernatural ones, from the saint (disease, disaster), and more concrete ones from the wifes kin: scorn, and the imposition of fines in the form of domestic animals that were to be sacrifices for the saint and whose meat was then to be distributed over the households of the village. Thus pilgrimage enabled to woman to maintain relations with her original segment (which she could hardly visit in any other context except pilgrimage), making her less dependent on her in-laws, and keeping alive her and her childrens claim on the estate in her original village.
In pre-colonial, nineteenth-century Khumiriya the worship of local saints was therefore a major factor in local integration. It allowed both for a manifestation of balanced opposition between brother-segments at various levels, and for the overcoming of this opposition on higher levels: directly by means of collective ritual and common identification with the higher level shrines, indirectly by its connection with the marriage pattern. Moreover the Khumiri religion in many respects reflected, reinforced and justified dominant notions and values in Khumiri society: the natural world, human life, human interaction, kinship, authority, etc.
Finally local integration was very much promoted by the political role of the guardians of the few major Khumiri shrines (minor shrines lacking such guardians).
Guardians succeeded one another according to a patrilateral adelphic system: upon a guardians death, his successor would be either his younger brother or, if he was the last of his generation, his eldest brothers son, etc. Guardians and their close kin distinguished themselves from the other Khumiris by avoiding violence and by leading what was locally considered a pious life. They would publicly observe one or several of the rules of conduct stipulated by the formal Islamic variant: perform the daily prayer, read the Quran, refrain from forbidden food and drink. Often these guardians were members of religious orders. The donations they received (especially on the saintly festivals they organized) were partly paid over to their superiors elsewhere. But even so guardianship formed a source of wealth, notably because of land that (as a donation to the invisible saint) was inalienably linked to the shrine. Because of their pacifism and their attachment to a fixed spot (the shrine), the guardians were more or less outsiders to the segmentary system. they were economically independent, and most effectively invulnerable because of their close association with invisible but powerful saints. therefore, the guardians were in a position to act as ultimate mediators between segments in case of important and violent conflicts that could not be resolved by the elders. Because of the guardians indispensibility for conflict regulation, secular groups could not allow the former to be harassed by other secular groups; by the same token, the guardians shrines were sanctuaries for refugees (especially in case of blood feud).
Many Khumiri shrines bear personal name, and are associated with personal, historical saints, about whom lively stories are told. so the history of these shrines may reflect, as we have seen, the history of the segments, but there are also individualizing, personal aspects involved. At least part of these shrines originated in the tomb of an historical holy man: one of the thousands of Islamic missionary agents who, mainly originating from Mauritania and Morocco, and as members of various religious orders, have flooded rural North Africa since the twelfth century.
In addition to a tomb, these holy men often left off-spring. Many names of clans in Khumiriya indicate a saintly founder: Ulad al Hajj, (sons of the Pilgrim"), Ulad ben Sayid (Sons of the Lord, i.e. Saint) etc. However, apart from a small guardian lineage, these name obviously soon lost their religious overtones; far from being pacifists, most bearers of these clan names fully participated in their violent segmentary society.
Several authors of maghribine religion have pointed out the ironical lot of these pious agents: gone out to replace local popular religion by their own version of formal Islam, they ended up with their tombs (transformed into local shrines) constituting the very corner-stones of local popular religion.
These holy men form an interesting case within the context of supra-regional integration. The formal Islam they represented had universalist tendencies beyond strictly local social structure, saints and shrines. The fact that the names of Khumiri clan-founders so often have religious specialist was a major entry for strangers who wanted to settle in this region. This was only possible if the Khumiris, despite their pursuit of a popular variant, regarded themselves as Muslims, identified with co-religionists elsewhere in the Muslim world, and welcomes Islamic specialists. however, this measure of universalism in the religious sphere was not accompanied by any supra-regional identification in other spheres, notably in politics. Oral data indicate that political identification hardly reached any further than the several tribal confederacies to be found in the Khumiri highlands. the agents of the central government (tax collectors) were violently kept out. Moreover, although universalism provided an entry for the pious agents, their teachings (to judge from the sate of Khumiri religion at the beginning of the colonial period) failed to bring about radial and permanent changes in the popular-religious notions and activities of the majority of the Khumiris. On the contrary, these formal-Islamic elements were neatly encapsulated and neutralized as the isolated status attributes of a very small minority of local religious specialists; who promptly were regarded as saints (i.e. inimitable); who had their functions not at the physical boundaries between Khumiriya and a wider structure but within the Khumiri segmentary organization (although, of course, they were straddling the cognitive, cultural boundary between Khumiri society and the outside world); and finally, whose initial links with supra-regional organizations (religious brotherhoods) must soon have become irrelevant, leaving barely recognizable traces in the memories and oral traditions of present-day Khumiris. Khumiriya was politically and economically isolated; Khumiri popular religion was closely connected with, and contributed very much to, Khumiri social organization; within this context there was little to promote and reinforce formal Islam on a larger scale. Probably, during the last few centuries Khumiriya never entirely lacked a handful of people who were able to read the Quran, and who observed a limited number of formal Islamic rules and prohibitions. However, the impact of these formal Islamic elements was very slight. Thus, for instance, until very recently Khumiriya never had its own real mosque; the larger local shrines were called mosque but were by no means places of collective weekly worship.
Whilst the pious agents did represent more formal and book-orientated versions of Islam than were prevailing in the Khumiri countryside, yet the versions they pursued (the notions and rituals of their own orders, emphasizing sainthood, ecstatic music and dance) were in many respects closer to Khumiri popular Islam than to the Islamic versions propounded by the urban theologians of their time. Forms of ecstatic ritual are said to have a history of many centuries in Khumiriya, possible as a local elaboration on what the pious agents brought. Nonetheless, the holy men formed a recurrent reinforcement of formal Islam in the Khumiri mountains. Because of them, Khumiri religion could develop in ever renewed contact with Islam elsewhere in North Africa.
4. The colonial Phase: Chief, Shrine and Brotherhood
although the French conquest of Tunisia (1881) was dictated by much wider political and economic considerations, it began as a punitive expedition against the Khumiris. Thus a long period ended in which in Khumiriya the influence of a central government had been minimal. The French stationed a garrison at the highest point of Khumiriya: the beginning of the town of cAin Draham. In 1889 they appointed local officials; they gave them the traditional title of shaykh (chief) and allotted to each of them an area (chiefdom) of several adjoining valleys. From cAin Draham, Khumiriya was brought under effective colonial control within a quarter of a century.
With the greater detail of historical data available on this period, our analysis will now centre on only a part of Khumiriya, the chiefdom cAtatfa.
Around 1800 a group of immigrants, and offshoot of a clan called cArfawiya after their apical ancestor cArfa, had arrived in this area and had since grown into a local numerical majority in two adjoining valleys. Of old, these immigrants appear to have been associated with the Shabbiya brotherhood. In one of the valleys, these cArfawiya founded (about 1850) the shrine of Sidi Mhammad, upon the tomb of a holy man from their midst. One of the cArfawiya families took to the guardianship of this shrine, which very rapidly became one of the most important shrines in Khumiriya. Some decades afterwards, in the course of short-distance migration, a branch of this shrine was erected a kilometre to the south.
About 1870 members of another line of descent within the cArfawiya clan, living in the other valley, founded there a lodge (zawya) of the Qadiriya order, after contacts with this order in Al Kaf. Many people in the cArfawiya and other clans joined the lodge as members (fuqra). In the first decades the lodge was rather effectively organized: in most of the villages it had representatives who annually collected donations, to be taken to Al Kaf by the founder-prior (muqaddim) of the lodge, and his assistants (shaush).
Betides their religious expansion, the economic expansion of the cArfawiya (who where renowned for their large herds of cAttle) became so great a threat to the inhabitants of two other adjoining valleys (mainly associated with the clans of Zaghaydiya and of Ulad al Hajj), that, about 1870, the latter allied to fight the cArfawiya. this alliance, headed by the Zaghaydi Yunis ben cAbul Qasim was to be the origin of the tribe of cAtatfa. The cArfawiya were weakened by the effects of an earth-quake, which also destroyed their lodge. Moreover their pacifist relatives, the guardians of Sidi Mhammad, exhorted them to end hostilities: dramatically the guardians cArried the saints sacred flags to the battlefield to determinate what was to be decisive battle. The cArfawiya then joined the alliance; their lodge was rebuilt a few kilometres to the east, on Zaghaydi land. Soon the tribal name passed on to the chiefdom established by the French.
After violent rivalry a Zaghaydi, brother of Yunis, finally was appointed chief by the French. He held this post until 1916, when he was succeeded by matrilineal kinsman (FBSS), also a Zaghaydi, but living in the valley of Sidi Mhammad (where the Zaghaydi meanwhile had come to be represented by a few homesteads). This second colonial chief was succeeded by his son (1939-57), so that for over forty years the cAtatfa chief lived in immediate proximity of the shrine of Sidi Mhammad.
Until the beginning of this century the chiefs power was still very limited. There were two main reasons for this. The French, his overlords, still did not yet control the region completely; moreover conflict regulation (formally the major task of the colonial chief, besides tax collection and the enforcement of other laws) was actually still in the hands of the wealthy guardians of Sidi Mhammad, who enjoyed a great religious authority. Moreover the chief was opposed by the lodge, whose prior and assistants had considerable economic power and, more important, great religious authority (not just among the lodge members, but among all those who consulted them for divination and healing: virtually all families in the wide environment). The closer relationship between guardians and the prior (both by common clan-affiliation, and by common association with the Qadiriya order) contributed to the initially weak position of the chief.
Gradually however the cAtatfa chief gained terrain on this religion-political complex.
He succeeded in building up a considerable economic power. Chiefs were not paid (until 1924), but were entitled to a share of the taxes they collected. Thus they rapidly acquired wealth. On the other hand, as elsewhere in the colonial world, the economic situation of the majority of the population deteriorated. The government began to exploit the cork-forests, restricted the making of new clearings and stimulated the establishment of a few expatriate-owned farms. Thus the agricultural area available for Khumiris diminished greatly and a stop was put to the segmentary dynamics. Tents gave way to huts and finally to stone houses; the villages were consolidated upon their present places. Land scarcity, erosion, rapid population growth (only to a limited extend encountered by migration out of the region), and a plague-plague (about 1930), increased the economic discrepancy between the chiefs and the great majority of the people. Besides the chiefs, only a small minority were able to profit from the rapid rise of cAin Draham as a regional centre and tourist resort. finally in the second quarter of this century the chiefs used their wealth and power in order to deprive a number of their fellow-villagers of their land rights. His increasing economic power made the chief enter into direct patronage relationships with many other heads of household.
The chief endeavoured to build his actual power (based on the colonial government and on his own wealth) into authority, by legitimizing this power in terms of dominant Khumiri norms and values: notably by the pursuit of the status of elder (kabir) as defined within Khumiri culture. By means of a worthy style of living, dyadic exchange relationships with many people, strategically chosen marriage ties, and conspicuous hospitality, the chief manipulated a local cultural idiom so as to enhance his power and to diminish local resentment of his person and position. It was in his defined-defined position of elder, no less than as an (externally defined) government official, that the chief gradually acquired the monopoly of conflict regulation. Along the same lines the chief tried to extend his authority into the religious sphere, where he sought to break the power of the religion-political complex of the cArfawiya, while remaining within the context of local notions of authority. These notions resolved on the interrelation between secular honour (ihtiram) and sacred grace (baraka): an elder should coordinate the interaction in his social field, including interaction with the sacred (collective ritual); and he should have optimal relationships not only with human beings, but also with non-human agents: saints. this is precisely what the chiefs tried to achieve.
Laws from the very beginning of the colonial era shows that in his struggle the chief was backed, to some extent, by measurements taken by the colonial government. Shrines were no longer acknowledged as juridical sanctuaries (1884), and land that had always been associated with shrines and orders, was declared alienable. We need further research on the impact of these general measures in the remote Khumiri highlands. Undoubtedly, however, these laws deprived the position of guardian of a supra-local official backing (which, on the contrary, did constitute the chiefs main power base). Guardianship became much less powerful and rewarding, and as a consequence, less attractive.
About 1920 the succession of guardians of the shrine of Sidi Mhammad underwent a radical change. From that time onward the local mens assembly has chosen the guardian among the close bilateral kindred of previous guardians. Whereas in the earlier patrilateral-adelphic system succession was entirely determined, the new system offered much more choice - while the chief influenced the final choice, and ratified the appointment. Thus in the years 1920-70 seven guardians succeeded one another who, through (mainly matrilateral) consanguinity with previous office holders, had acceptable (though not patrilateral-adelphic) claims to guardianship, but who, on the other hand, were very closely linked to the Zaghaydi clan and even (1940-70) were fully dependent clients of the chiefs.
In this way the chief acquired indirect but effective control over the most important shrine of his chiefdom. In the 1920s the chief personally took over the organization and co-ordination of the local saintly festival. In 1930 the created a new cemetery around the original shrine of Sidi Mhammad. The family of original cArfawiya guardians who until the beginning of this century locally dominated a number, wealth and power, fell into decay: by now all its male members have left the surroundings of Sidi Mhammad.
That saintly arbitration was a dying institution is well demonstrated by the fact that Khumiri oral history records no living saints after the 1910s. Unlike Morocco, in Khumiriya the saint today is always a dead saint.
With regard to the lodge we see the same striving for indirect chiefly control. Whereas until the beginning of this century the lodge-members living around Sidi Mhammad all belonged to the family of cArfawiya guardians, in the later decades nearly all local lodge-members belonged to the closed kindred of the chief. Locally, the latter-day lodge-members predominantly belong to recently immigrated families, bound to the chief by ties of patronage, and enjoying little prestige. This situation is exceptional as compared with other villages in the research area. Throughout the area nearly 20 per cent of the heads of households are lodge-members, and this number is sufficiently large to make a quantitative analysis possible. such analysis demonstrates that in other villages lodge-members do not differ from non-lodge-members, as to wealth, prestige, and the period of local residence of their matrilineal descent line. Unmistakably the lodge-members around Sidi Mhammad, whose social status is much lower than that of lodge-members elsewhere, are the chiefs pawns in his encroachment into local religion.
But while the chiefs control over shrine-guardianship has implications for the entire chiefdom, the chiefs indirect control over lodge-members has however been limited to just one village. In the course of about half a century the regional organization of the lodge has become increasingly loose. I suspect that his was partly due to the impact of the colonial government, which through the local chiefs may have tried to weaken such rival, local foci of power as the lodges represented vis-à-vis the central power. In addition, built-in structural tendencies within the local social field seem to have torn apart the lodges regional organization in Khumiriya. the religious brotherhood creates interaction and mutual identification between lodge-members irrespective of their segmentary distance in every day life. this is however contrary to the working of a segmentary system; and with segmentation forming the main structural principle in Khumiri society, it looks as if the cross-cutting regional lodge organization was gradually broken up by the localizing segmentary tendencies. In a nutshell this would be the same process as which on a larger scale, made for the proliferation of local lodges, shrines and saints throughout North Africa: as they became effectively incorporated in a local segment, religiously-based supra-localities *check this passage* (linking local offshoots to their remote parent establishments) became impedimental to these offshoots integration in the local segments ; as these offshoots survival would depend on local functioning, the supra-localities *check this passage* would become irrelevant, and wither away. Territorial segmentation appears to be conducive to local autonomy of religious foci (office-bearers, shrines), and inimical to religious supra-local organization such as the dispersed membership of the lodge represented.
At any rate, because of this looser organization, control over lodge-membership in one village does not give the chief any real power over the prior and his assistants, who live in other valleys than the chiefs. The chief has not been able to destroy the great authority of these religious specialists, and conflict between chief and lodge has endured until today.
5. Enhancement of Formal Islam
The French conquest of Khumiriya initiated an enhancement of formal Islam in this area.
In the beginning of this century, some Quran teachers settled in Khumiriya from abroad (Morocco; Kabylia). Half a century later most formal Islamic knowledge of most villagers (distinct notions about god, the Quran and afterlife; chants about the Prophet; myths about the major North African saints) cAn still be traced to these teachers. In principle their action was a continuation of that of the pious agents in pre-colonial Khumiriya. In the new situation, however, formal Islam appears to have a novel effect. It creates a permanent alternative perspective that directly threatens Khumiri popular religion. Are the Khumiris true Muslims? cAn the local saint really take revenge, cAn he help in times of illness and distress? Are sacrifices of animals and other food really pleasing to god? Is the lodge-members ecstasy a mystic union with god, His Prophet and his saints, or is it a diabolical cult, or even merely a kind of sport, or a conjuring trick? Is the worship of shrines other than the few most important ones acceptable? Are cArd-playing, drinking and eating wild pig perhaps most more serious sins than the general Khumiri indulgence in these activities suggests? These are some of the questions which few Khumiri supporters of formal Islam now raise concerning the local version of popular Islam.
Undoubtedly, through the centuries, the recurrent confrontation with pious agents temporarily raised similar questions. But, obviously, the ensuing doubt then gave way again to the notion of essential continuity between formal Islam and the Khumiri popular version. The then greatest threat to this notion of continuity - the pious agents themselves - would be dispelled by making the latter into the cornerstones of the local social structure and popular religion.
In the colonial age, on the contrary, an enduring notion of discontinuity was to emerge. Certain Khumiris were confronted, in cAin Draham and elsewhere, with formally-Islamic urban Tunisians, and with the French authorities who, in religious matters, based their policies largely on what they considered to be formal Islam. Adopting formal-Islamic elements was attractive, or sometimes even necessary, for those Khumiris who had vested interests outside the Khumiri villages, and who, in this cApacity, occupied the strategic positions in the relation between the rural villages and wider structures: Khumiri chiefs, their assistants and close relatives, traders (in cAin Draham), the (very few) pupils of primary and secondary schools, and professional soldiers after their retirement. These cAtegories were to form an incipient Khumiri elite of relatively powerful and wealthy people. In addition to western clothing and housing, literacy, the French language and smoking of cigarettes, formal Islamic elements became the status attributes of this elite. As such, these elements were also adopted by a few social and economic climbers among the villagers who, as yet, lacked the important supra-regional contacts but for whom the elite formed a reference group.
Already in the 1930s Demeerseman noted the emergence of this new formal-Islamic perspective. The decline of mass saintly festivals, which set in by that time, is partly explained by the economic decline of the majority of the population, and by government prohibitions; but addition the growing impact of formal Islam forms a major factor. What was historically the ritual climax of these festivals, the collective meal, has disappeared completely after the period 1935-50. Some visible manifestations of the same process occurred already much earlier. In the beginning of this century the few most important Khumiri shrines (including Sidi Mhammad), until then simply huts constructed of rough stones and arboreal material, were transformed into white stone buildings roofed by a dome (qubba). The later type of shrine is found throughout the Islamic world, and is far more acceptable to formal Islam. Significantly, Khumiris at that time lacked the required skills and had their early qubbas built by Europeans.
Apart from this reshaping of already existing shrines, from about 1925 onward no entirely new shrines were created any more. Thus the relatively young village of Hamraya has no man-made shrines. From the same recent period we have the first and only reports of Khumiri shrines being demolished. One very important shrine, although not a qubba (yet), was destroyed by an expatriate settler about 1920: he built his farm upon it. In the 1940s, another shrine, small and forgotten, was discovered and destroyed by a pious Khumiri, a retired professional soldier, who was making a new clearing.
6. Formal and Popular Islam Today
Economic decline befell Khumiriya during the colonial era, and largely as a result of the colonial situation. The decline continued during the first decade after independence (1956), and necessitated unemployment relief work to be organized. The government created a large re-afforestation project; however much this project may contribute to the future of the region, its direct effect has been that the agricultural area available for private peasants became still more limited, and the goat husbandry was prohibited. Misery and distress are paramount, and the general attitude vis-à-vis the government, the national party, and national goals is negative. Rudebeck remarked that the rural proletarians of Tunisia still do not appear to be integrated into the political system, if this is read to mean that the peasants have no active part in, nor motivation towards, shaping the economic and political processes which affect them so dramatically, the remark certainly applies to Khumiriya; (of course, the powerless peasants have little choice to be integrated in the political system as passive objects of policy). The Khumiri situation closely resembles the one Duvigneau described for the village of Shebika, some 120 kilometres south of Khumiriya.
Meanwhile the growth of local elite has continued. They developed their own style of living, with the status attributes already mentioned, and including the tendency of elite members to associate, for daily interaction and marriage, preferably with one another. they acquired influential positions after independence, not only as chiefs and their assistants, but also in local party organization and the relief work organization.
Formal Islamic elements continued to increase in importance. Formal Islam not only provides status attributes, but also forms a channel of upward social mobility in the non-religious sphere. It renders socio-economic climbers in the village acceptable associates, though not equals, of the elite. Moreover there are several recent cAses of Quran teachers (both in the villages and in the new mosque ofcAin Draham) who, through the prestige and the network of relations built up as religious specialists, acquired prominent posts in local government or in the relief work organization - promptly to resign as religious specialists.
The pull of status advancement is also strongly felt among religious specialists in the popular variant. These too now tend to give up their religious offices without hesitation as soon as they get a change to build up (often on the basis of their religious specialism, in combination with other skills, network relations, inheritance, etc.) some wealth and power in the non-religious sphere. this process involves only a few persons, because opportunities are scarce, and because numerically specialists only form a minority. However, several of todays richest and most powerful villagers are former lodge-members, who left this religious specialism to their less fortunate fellow-villagers. Moreover some musicians, specialized on the instruments for ecstatic sessions (where they are not supposed to make money) have changed to the instruments for festive music, which earns them an occasional but considerable income on festivals, weddings and circumcision ceremonies. Already a quarter of a century ago, even a guardian of the shrine of Sidi Mhammad resigned from guardianship to pursue festive music as a remunerative part-time specialism.
The present situation is confusingly ambivalent in various respects. Due to the spread of literacy (partly through private, small Quran schools in the villages), nowadays many villages under thirty years of age are able to read the Quran: at burials, memorial rites, circumcision and wedding ceremonies and at the general Islamic festivals. The readers derive prestige from their awe-inspiring ability. Yet for illiterates the classical Arabic text remains practically incomprehensible when recited. And the teachings of the Book by no means eclipse the popular Khumiri version of Islam. People who resigned from their specialisms in the sphere of popular religion take pains to deny or to excuse this fact. The elite (and those who aspire to belong to it) emphasize formal Islam as a status attribute. But at the same time they keep participating in the popular variant: they still feel best at home in the latter sphere, and in their view the responsibility, envy and conflicts connected with their exalted position necessitate optimal relations with the powerful and legitimacy-providing local saints. The popular, saintly religion instrumentally provides the local elite with the opportunity to show its political and economic power and to build up authority: by means of impressive animal sacrifices; by organizing the construction of a few qubba; by trying to acquire official permission for a saintly festival (which in view of prevailing official attitudes is no mean task); and by actually organising such a festival. Moreover some members of the elite make a substantial profit by trading at the great saintly festivals.
North African religious festivals have a redistributive aspect which may be relevant in this context. The sacrifice of a domestic animal dedicated to a local saint is a major element in the popular religion. The meat cAnnot be sold but has to be shared out among local people and passers-by. In a small-scale society where domestic animals are relatively abundant, where differences in wealth are moderate and where reciprocity dominates social life, such ritual redistribution may serve to convert dispensable wealth into honour at low extra costs. In contemporary Khumiriya however cAttle and sheep have become scarce and mainly concentrated in the hands of the elite. Wealth differences are great. The elite cAnnot expect reciprocity from the village poor. Nowadays the elite lets the increase of wealth prevail over the increase of honour. In this context animal sacrifices have grown too costly for the elite. Since deep involvement in the popular sphere implies animal sacrifices for those who cAn afford it, the elite is induced to shun this sphere and instead to pursue the formal variant: animals killed on the Prophets festivals are not supposed to be distributed gratis. Thus the elite continually oscillates between indulgence in and rejection of the popular variant.
In most local contexts both Khumiri elite and non-elite tacitly presuppose essential continuity between formal Islam and local popular Islam. Notions out of both variants are intermingled without differentiation, and one ignores such considerable theological problems as have been mentioned in the beginning of section 5. This implicit assumption of continuity could be interpreted as an unconscious attempt, on the part of the Khumiris, to escape the cognitive dissonance between on the one hand formal Islam as a local ideal reinforced from outside and on the other popular Islam as a fondly cherished local practice. But beyond this, the participants assumption of continuity primarily bears out the central historical fact that for centuries the two variants have been at dialectical interplay within Khumiriya. Both are part and parcel of Khumiri society, and if anything cAlls for a sociological explanation, it is not the tolerance between them in most situations, but the intolerant polarization in some.
The painful situations in which this convention of continuity is explicitly rejected, contain valuable hints as to the fundamental political and economic processes underlying the tension between formal and popular versions of Islam in contemporary Khumiri society. One situation in which the popular variant finds itself explicitly rejected is when district authorities (typically non-Khrumirs) occasionally refuse permission for a popular collective ritual.
Since independence the Tunisian government, in its quest for modernization, has opposed popular Islam. Permission for saintly festivals were refused, even, outside Khumiriya, a number of shrines have been demolished by the authorities. Certain elements in the ritual of the religious orders were prohibited, and the orders were severely criticised. In Khumiriya, timing and scope of the annual saintly festivals are now strictly controlled by district headquarters, whereas the ecstatic ritual is officially forbidden (especially the most spectacular elements of it: manipulation of fire and knifes during trance).
Rejection of the popular variant also occurs when Khumiri members of the ritual elite refuse to participate in customary popular activities such as the ecstatic dance, the worship of minor shrines, or the consumption of wild pig. Formal Islam provides them with a standard argument against these activities; they are cAlled: haram, i.e. prohibited by Islamic law, hence polluted, taboo. In the village, such rejective behaviour is often triggered by the presence, as a third party, of townsmen: people who are considered representatives of the ideal, formal version of Islam, and who (in this, and in most secular respects) from the reference group par excellence for the rural elite. The ostentatious rejection of local religious forms by elite-members who happen to be present, creates strong negative feelings in their non-elite fellow-villagers. The latter regard it as a manifestation of assumed superiority, and in addition feel threatened in their assumption that their popular religion represents true Islam. The elites rejection destroys the assumption of formal/popular continuity, and forces the underlying cognitive dissonance into full consciousness.
Confusion and embarrassment are the peasants dominant responses to this confrontation. But on the other hand, the formal variant of Islam, and its local exponents, cAn be heard to be explicitly rejected by Khumiri peasants when discussing, with scorn and resentment, the alleged machinations of local authorities and elite both within and outside the sphere of the religion. Under modern conditions, the pursuit of popular collective ritual (particularly the ecstatic sessions and the saintly festivals, even if in a greatly reduced form) is not just a continuation of custom; rather it takes on unmistakable aspects of an awakening local consciousness in terms of ethnicity and class. As my informants put it: Whatever "they" may say or do, this is what we, people of the mountains, have always done and will continue to do.
Whereas many Third-World scenes now show a decline of historical forms of local religion, the Khumiri response is rather that of a revival.
Despite the economic decline, recent years have seen a trend to reshape existing shrines into qubbas - an expensive and labour-consuming task in which whole villages co-operate. Each family is prepared to pay its share for the festival permission. When an official permission was sure to be refused, ecstatic sessions were held clandestinely. In the new generation, though now largely exposed to primary-school education and long-term military draft outside Khumiriya, many new lodge-members are being recruited. It is not only the elderly, but also young people and children who emphasize their relationship with the local saint, and spontaneously visit his shrine. With territorial segmentation still the major social-structural feature, saints worship continues to provide the shrines and cemeteries to mark the segments at various levels.
Thus it would appear as if, recently, the popular and the formal aspects of Islam in Khumiriya are becoming polarized in a way which is rather at variance with their local intertwinement over past centuries. The explanation seems to lie mainly in colonial and post-colonial processes of political and economic incorporation. The colonial situation had dislocated the loci of decisive political and economic power to far outside Khumiriya and had reduced the average Khumiri to the status of unemployed-relief worker or frustrated would-be labour migrant. For those who, as members of the local administration and the rural elite, act as intermediaries between this outside power and the powerless peasants, Islam has become an aggressive status attribute which when brought to bear upon face-to-face interaction between elite and peasantry, enables the former to demonstrate their derived power and to emphasize their relative independence vis-à-vis local, popular custom - at the same time confusing and provoking the peasants. For the peasants the embarrassment and resentment instilled by the elites religious challenge is just one aspect of their general position of powerlessness. Under the present circumstances, they lack the economic and political power, the analytical understanding of their predicament, and the secular organization, to effectively adopt other symbols (let alone actions) for the expression of discontent; therefore entrenchment in the popular religion has become the main (though not the only) expression of their predicament. The theoretical problem of why and through what mechanism, here as elsewhere, a societys political and economic transformation had to find a predominantly religious expression, is something beyond the scope of the present paper.
The tensions of the situation pervade all aspects of contemporary Khumiri religion. However, in certain cAtegories of people these tensions are particularly acute and give rise to profound social and psycho-somatic crises; this is the cAse with resigned or still active specialists in the popular variant; members of the elite; and the close kinsmen of these people. This involved try to resolve these crises, not through political action based on macroscopic analysis of economic and political incorporation, but in terms of exceptionally severe conflicts with local saints in the idiom of ecstatic cults of affliction.
In this paper I have attempted to show how in the religion of Khumiriya (an area not a-typical for much of rural North Africa) the coexistence of a popular and a formal version of Islam, and the various forms this coexistence has taken over time,cAn be profitably (though by no means exhaustively) analysed in terms of supra-local political and economic relations. In the centuries when Khumiriya was a segmentary society with very limited political and economic relations with the outside world, the perception of this outside world as sharing in essentially the same religion, provided for contacts with pious strangers performing (in addition to economic function which I have not the data to discuss systematically) crucial political functions within Khumiri society. The fact that these pious strangers could be assimilated locally without substantially affecting local political and economic power structures, allowed for the religious variant these strangers represented (i.e. formal Islam) to be organically accommodated, serving the local needs for outsider arbitration. In his studies of the Moroccan High-Atlas, Gellner has demonstrated how there the saints mediate both between segments within the local rural society, and between that society and the outside world: coastal Morocco, the colonial government, the colonial government, the world of Islam. The present study applies a similar view to a different part of North Africa.
Colonialism (and the perpetuation, after Independence, of the structures it had created) dramatically upset the pattern of supra-local relations surrounding Khumiriya. The focus of effective political and economic power shifted from local communities to modern bureaucratic organizations outside Khumiri rural society. Among the local agents (administration, elite) of the new power structures, the emphasis on formal-Islamic elements precipitated a polarization between formal and popular Islam in the area. In the present situation the historical intertwinement between the two variants of Islam has largely become eclipsed, and (as a starting point for a more secular peasant movement?) popular Islam has developed into a major symbol of growing peasant consciousness. Meanwhile one wonders what effects an improvement of the local economic and political situation will have on
the relations between the two versions of Islam in the area.
 Data were collected during fieldwork in 1968 and 1970. The ethnographic present refers to the late 1960s. I am indebted to the following persons and institutions: the University of Amsterdam, which provided a research grant and under whose auspices D.G. Jongmans established the Khumiriya project within which I carried out the fieldwork; the Centre des Arts et Traditions Populaires, Tunis, which co-ordinated the project locally; D.G. Jongmans, who generously shared his profound knowledge of the region and who guided the project throughout; Hasnawi b. Tahar, for excellent research assistance; H.E. van Rijn, my wife for sharing my prolonged interest in Khumiriya, and for guidance in the quantitative analysis upon which the present argument is partly based; J.F. Boissevain, M.L. Creyghton, E. Gellner, A.M. Hartong, A.Huitzing, H.J. Simons, K.W. van der Veen, P.C.W. van Dijk and J.M. Van der Klei, for helping discussions on parts of the argument; and finally O. Mwelwa, A. Schijf and M. Zwart for typing various versions of the manuscript. Earlier versions of this paper were presented at a seminar of Zambia and at the Universities; Social Science Conference. Nairobi, both in December, 1971.
 Cf. E.C. Hagopian, The Status and Role of the Marabout in Pre-Protectorate Morocco, in: Ethnology, 3, 1964, p. 42-52.
 Westermarck, E., Marriage Ceremonies in Morocco, London, 1914; idem, Ritual and Belief in Morocco, I.& II, London, 1926; Montagne, R., Les Berbères et le Makhzen dans le sud du Maroc, Paris, 1930; idem, La vie sociale et la vie politique des Berbères, Paris, 1931; Evans-Pritchard, E.E., The Sanusi of CyrenaicA, Oxford, 1949; Peters, E.L., The Sociology of the Bedouin of Cyrenaica, D. Phil thesis, Oxford University, 1951; Berque, J., Structures sociales du Haut Atlas, Paris, 1955
 Geertz, C., Islam Observed: Religious Development in Morocco and Indonesia, ChicAgo & London, 1968.
 E. Gellner, Saints of the Atlas, London, 1969; Van Binsbergen, W.M.J. Religie en Samenleving: Een studie over het Bergland van Noord-West Tunesië, Drs. Soc. Sc. thesis, University of Amsterdam, 1971, idem, Verwantschap en Territorialiteit in de Sociale Structuur vn het Berland van Noord-West Tunesiè, Drs. Soc. Sc. thesis, University of Amsterdam, 1970; a combined English version of these studies is currently being prepared.
 Gellner, o.c.
 Mason, J.P., Saharan Saints: Sacred Symbols or Empty Forms?, in: Ethnology, 13, 1974, p. 390-405.
 Creyghton, M.L., Folk Illness in een Tunesisch Dorp, Drs. Soc. Sc. thesis, University of Amsterdam 1969; Crapanzano, V., The Hamadsha: A Study in Moroccan Ethno-psychiatry, Berkeley, 1974.
 Van Binsbergen, Religie, o.c.
Köbben, A.J.F., Opportunism in Religious Behaviour, in: Van Beek, W.E.A., & Scherer, J.H., eds., Explorations in the Anthropology of Religion, The Hague, 1975, p. 46-54; the quote is on p. 50.
Gellner, op. cit.; idem, Sanctity, Puritanism, Secularism and Nationalism in North Africa, Archives de Sociologie in a Pilgrimage Center, Austing & London, 1976.
 Bel. A., La religion musulmane en Berbérie, Esquisse dhistoire et de sociologie religieuses. I, Paris, 1938: Draque, G., Esquisse dhistoire religieuse du Maroc: Confréries et Zaouîas, Paris, n.d. (1951); Trimingham, J.S., A History of Islam in West Africa, London, 1962, p. 16f.
 Van Binsbergen, W.M.J. Saints of the Atlas: Ernest Gellner", in: cAhiers des Arts et Traditions populaires, 4, 1971, p. 203-11.
Hartong, A.M., De Geschiedenis van het Sjeikaat Atatfa op Basis van de Orale Traditié, Drs. Soc. Sc. thesis, cAtholic University of Nijmegen, 1968; Souyris-Rolland, M., Histoire traditionelle de la Kroumirie, in: IBLA, 12, 1949, p. 127-65; Demeerseman, A., Le culte des Walis en Kroumirie, in: IBLA, 27, 1964, p. 119-63; Ling, D.L., Tunisia: from Protectorate to Republic, Bloomington & London, 1967; Van Binsbergen, Religie, Verwantschap, op. cit.
 Cf. Fortes, M., 7 Evans-Pritchard, E.e., Introduction, in: idem, African Political Systems, London, 1940; Middleton, J., & Tait, D., eds, Tribes without rulers in African Segmentary Systems, London, 1958.
 Favret, J., La segmentarité au Maghreb, in: LHomme, 6,1966, p. 105-111: Gellner, Saints, op. cit.
 Peters, E.L., Some Structural Aspects of the Feud among the cAmel-herding Beduins of Cyrenaica, in: Africa, 37, 1967, p. 261-82.
 Van Binsbergen, Religie, Verwantschap, op. cit.
 Souyris-Rolland, op. cit.
 In the present paper I am concerned with the social structural, as analyticAlly distinct from the cultural, dimention of Khrumir religion. my research has however entailed both aspects and attempted to combine them. this had led me to an analysis of such dominant themes as honour (ihtiram), grace (baraka), sainthood, and man-saint relaitonships, against fthe general cultural and social-structural background. I am fully aware of the great importance of these aspects for an understanding of Khrumir society and religion, yet in the present argument will only cursorily deal with them.
 Cf. Gellner, Saints, o.c.
 This is not to suggest that foreign origin was a necessary condition for posthumous sainthood. Throughout North Africa, locals have been known to become considered as saints after their deaths. However, in Khrumiria pious legend has invariably associated such local saints not only with wonder-working and an exemplary social life, but also with elements derived from the formal, outsider version of Islam: reading of the Quran, regular praying, pilgrimage to meccA, the white burnouse, the observance of food prohibitions etc. However local and popular the saint may have been while alive, once he becomes cAnonized locally he is conceived mainly in the trappings of a pious stranger.
 Gellner, Saints, o.c. Geerts, o.c.
 By contrast, in the Moroccan Atlas, living saints and the saintly tombs they administer straddle the boundaries between major local groups, their territories, and between major ecological zones; Gellner, Saints, o.c.
 The ecstatic element in North African religious ordrs is said to derive from three sources: early Islam in the Middle East (cf. Molé, M. La Danse extatique en Islam, in: Les Dances Sacrées, Anthologie, sources Orientales, Paris, 1963, p. 145-280); ecstatic cults derived from sub-saharan Africa (cf. Brunel, R., Essai sur la Confrérie religieuse des Aissaoua au Maroc, Paris, 1926; Trimingham, J.S., Islam in the Sudan, London, 1965); and autochtonic ecstatic cults dating back to Antiquity (cf. Bertholon, L. & Chantre, E., Recherches anthropologiques dans la Berbérie orientale, I, Lyon, 1913). In addition to anthropological studies, much more historical research is needed on this point, such research could greatly benefit from the recent theoretical, methodologicAl and factual advances made in the field of pre-colonial religious history of sub-saharan Africa, e.g. Ranger, T.O., & Kimambo, I., eds., The Historical Study of AfricAl Religion, London, 1972.
 Ling, o.c. p. 22f; Abun-Nasr, J.M., A History of the Maghrib, 2nd ed., London, 1975, p. 278f.
 Originally they belonged to the Drid tribe. cf. souyris-Rolland, o.c., p. 135; Bel, o.c., p. 378f; Hartong, o.c., p. 36; Miedema, A.W.F., Verslag Leeronderzoek Tunesiè 1965, typescript, University of Amsterdam, Anthropologisch-Sociologisch Centrum, 1967, p. 19; Cuisenier, J., Endogamie et exogamie dans le marriage arabe, in: LHomme, 2, 1962, p. 80-105; Van Binsbergen, Verwantschap, o.c., p. 93f.
 Al-Kaf is a town about 100 km south of Khrumiria. The Qadiriya order was founded in the 12th century by Sidi Abd al-Qadir al-Djilani, in Baghdad; since the order has grown to be one of the most important and widespread orders of entire Islam; cf. margoliouth, D.S., Qadiriya, in: Gibb, H.A.R., & Kramers, J.H., eds., Shorter Encyclopaedia of islam, Leiden, 1974, p. 202-5, and references cited there.
 Van Binsbergen, Religie, Saints, o.c.; cf. note 20.
 These laws were enacted in 1889; cf. Ling, o.c. p. 59.
 Van Binsbergen, Religie, p. 208f, 295f.
 For Central_AfricAn developments reminiscent of this same process, and for a more general discussion of the regional dynamics involved, cf. Van Binsbergen, W.M.J., Explorations into the History and Sociology of Territorial Cults in Zambia, in: Schoffeleers, J.M., ed., Guardians of the Land: Essays on Central-African Territorial Cu7lts, Gwelo, 19790, idem, Regional and Non-regional Cults of Affliction in Western Zambia, in: Werbner, R.P., ed. Regional Cults, A.S.A. Monographs 16, London, 1977, p. 141-175; also cf. Werbner, R.P., Introduction, in: idem, Regional Cultus, op. cit, p. ix-xxvii.
 Jongmans, D.g., Meziaa en Horma: Samenhang tussen Dienstbetoon, Eer en Welstand in een Veranderde Samenleving in: Kroniek van Afrika, 3, 1968, p. 1-34; Van Binsbergen, Verwantschap, o.c. p. 75f.
 As recApitulated in his 1964 article, o.c.
 Rudebeck, L., Party and People: A Study of Political Change in Tunisia, 2nd ed., London, 1969, p. 265.
 Duvigneau, J., Chebika, Paris, 1968.
 Standard musicAl instruments for ecstatic sessions are the kusba (flute) and the bendir (tympanon, a large tambourine without bells); typical festive instruments are the tabbala, a large and high drum played with sticks, and the zukkra, a kind of hoboe; cf. Van Binsbergen, W.M.J., Muziek en Dans in het Atlasgebergte, in: Muziek en Volkenkunde, nos. 109-10 and 111-2, 1971.
 Jongmans, op. cit.; idem Politicts on the Village level, in: Mitchell, J.C., & Boissevain, J.F., eds., Network Analysis; Studies in Human Interaction, The hague/Paris, 1973, p. 167-217.
 MicAud, C.A., Social and Economic Change in: idem, Tunisia, The Politics of Modernization, London, 1964, p. 144.
 Speight, R.M., Tunisia Sufism;, in: Moslem World, 56, 1966, p. 58.
 Cf. Van Binsbergen, W.M.J., Religious Change in Zambia, London/Boston 1981, esp. chapter 1.
 This article was prepared for publicAtion in the course of my current employment with the African Studies Centre, Leiden, The Netherlands
|page last modified: 13-02-01 08:43:40|