Caste is a form of social differentiation characterized by “pollution”, with one group seeing another as “dirty”, “smelly”, or some other term used in a metaphorical sense to differentiate groups existentially and hierarchically: the clean are always considered superior to the dirty. This leads to limitations in social relations, marriage for example being forbidden between members of the clean and the unclean groups. In such a case anthropologists describe the castes as endogamous, meaning that one marries within one’s caste. In some parts of the world, notably in India, there are several castes and subcastes arranged in a hierarchy of pollution from the “ritually pure” Brahmins to the Dalits, formerly “untouchable”. Each caste group is there associated with its own range of economic specializations

Caste in the Mandara Mountains

Caste is a form of social differentiation characterized by “pollution”, with one group seeing another as “dirty”, “smelly”, or some other term used in a metaphorical sense to differentiate groups existentially and hierarchically: the clean are always considered superior to the dirty. This leads to limitations in social relations, marriage for example being forbidden between members of the clean and the unclean groups. In such a case anthropologists describe the castes as endogamous, meaning that one marries within one’s caste. In some parts of the world, notably in India, there are several castes and subcastes arranged in a hierarchy of pollution from the “ritually pure” Brahmins to the Dalits, formerly “untouchable”. Each caste group is there associated with its own range of economic specializations.

In the Mandara mountains area, the montagnards recognize only two castes, which we can call the “farmers” and the “specialists”. This is not to say that farmers do not engage in specialist activities, for example hunting, basketry, divination or constructing mud domes over round rooms, nor that specialists are forbidden or choose not to farm. However, among the montagnards the work of forging iron is always the preserve of male specialists and making pottery is very commonly that of their wives and daughters, for which reason specialists are often described in the ethnographic literature as smiths, blacksmiths or smith/potters. And while all farmer men were expected to serve as warriors in times of conflict, specialists were sometimes excused such duties and treated as non-combatants. Not all montagnard groups are casted: the Mofu-Diamaré for example are not, though they share many features with their casted neighbors and among them marriage of non-smiths to smiths is discouraged. There is thus a gradient between casted and uncasted montagnard societies.

Montagnard farmers regard specialists as unclean and, although the latter do not necessarily or fully concur with this characterization, both castes accept their differentiation. This is not expressed in a marked degree of hierarchy and is commonly explained as originating as a differentiation between brothers, either resulting from their different talents, or from a polluting action taken in the distant past, for example one brother eating a squirrel or other “bad” animal, or alternatively a “good” animal sacrificed at a funeral with hands unwashed since tending the dead body. The former explanation is more likely offered by specialists, the latter by farmers. The familial relationship is also implicit where a clan, like the Rəvai of Sukur, includes members of both castes. At Sirak, south of Mokolo, during Gawula, the ceremony that initiates married men into elderhood, ancient brotherhood is expressed by specialists and farmers drinking out of the same calabash, something they normally avoid. The brotherhood myth takes various forms, some involving one brother tricking another, but the result is the same: one brother becomes a specialist differentiated from the farmer majority. Specialists are few compared to farmers, ranging in former times when montagnards produced their own iron from about 1% up to around 12% among those groups most specialized in iron metallurgy. 

The differentiation of montagnard groups into two castes has important economic implications. Anyone can sharpen an iron tool or make a toy or small bowl out of clay, but it is quite another matter to smelt, fine and weld iron from local ores and to fashion it into the full range of tools, or to create a complete set of pot types by digging and processing clay, pot-forming, decorating, drying, firing and eventual distribution of types that vary from personal eating bowls to huge storage vessels, and which include figurated forms designed for social and ritual purposes. Both metallurgy and ceramic production require specialist knowledge, techniques, management and marketing skills beyond those that can be acquired by someone who has comparably but differently invested his or her time in learning the botany, pedology, livestock breeding and management, crop processing, storage, and other skills required of a farmer, let alone one who must maintain his ability to defend his house or home, or in the case of women manage other aspects of the farmstead.

The two caste system does not ensure that all political entities, such as Mafa villages, or even all ethnic groups are self-sufficient in iron artifacts and pottery, both critical factors of production. Montagnard communities were often themselves specialized and, even before locally produced iron was replaced by industrial stock and scrap from Europe and elsewhere, traded in a considerable number of commodities. Trade and exchange together with the caste system ensured that the Mandara montagnard community as a whole was supplied with the metal and ceramic artifacts that they required, and that specialists in every settlement could maintain toolkits throughout the year. For example there is a high demand for repairs and sharpening of agricultural tools in the late dry season and the first half of the rains. Housewives try to acquire a year’s supply of pots before the dry season ends, but if a household finds it needs a ritual vessel in the height of the rains, there will be a potter who can supply it.

Seasonality is an important factor in montagnard life. Iron smelting was carried out in the later dry season (March – early May) after the main sorghum/millet harvest was completed and the grain threshed and stored away. The work of the forge continued throughout the year at varying intensity. Almost all pot manufacture took place in the dry season when it was relatively easy to dry vessels before they were fired, thereby avoiding the cracking, flaking and breakage during firing resulting from the explosive transition to steam of water retained in vessel walls.  

The combination of the requirement of technological expertise with seasonal demands on smith/potter labor has interesting consequences. In particular both male smiths and their female potter counterparts, who often assist at the bellows their men in the forge, are constrained in their ability to cultivate crops during the rainy season, a time when weeding consumes a great deal of farmers’ time and effort. Similarly they have less to do at harvest time than farmers. The seasonal round thus discourages smith/potters from farming while offering them considerable interstitial time that they are well-equipped to fill with other specializations. These include curing (botanical knowledge is important both in medicine and smelting), midwifery, funeral direction, divination, magical removal of foreign objects from the human body, leather and wood-working, and musicianship, among several others. While smith/potters typically monopolize the work of the forge and ceramic production, they and their communities make different selections among other specializations, which they typically carry out in varying degrees of competition with farmers.

In the social realm, the brotherhood metaphor often repeats itself politically in a complementarity between the chief and the chief smith, where both are required in order properly to carry out important divinations and ceremonies.

Funeral direction is a specialization that has fascinated European observers to the point that some have argued that it is dealing with the dead that renders smith/potters unclean. An alternative view is that smiths are required, sometimes forced, to act as morticians because they are unclean and few. However, many disposals of the dead have always been undertaken by relatives of the deceased, those of infants, children and lepers for example, demonstrating that handling dead bodies did not in itself necessarily pollute. Often a sister’s children were and are called upon to conduct funerals of members of their maternal uncle’s family (Fig. 1). In recent times smith-directed funerals are often limited to those of elders. Burial ceremonies for important persons commonly extend over several days and are likely to include a smith carrying a several-day-old decomposing corpse seated on his shoulders, “dancing with the dead”. For this unpleasant service they will be compensated with a good share of the meat of the animal(s) sacrificed for the occasion and other goods.

Figure 1. At a Sukur funeral a maternal nephew carries the body of an old female relative on his shoulders (17 Feb. 1993).

Caste at Sukur

Note Like all pages on this website, the authors’ “ethnographic present” is the period 1991-96 during which we did the bulk of our fieldwork. Of course we learnt more on other visits and from others’ accounts of Sukur and other Mandara montagnards. But the subject of this page, caste and particularly the smith/potters or ɗai, is one that is sensitive to changing circumstances, thus we are describing a social institution as it was during a particular period.

Demography and Clans
The two castes, farmers (mbəlim ) and smith/potters (ɗai), are present among the Sukur, and caste structures important aspects of life. In early 1993 there were 212 Sukur households resident on the mountain in upper Sukur of which 15 (7.1%) were ɗai, the others being farmers. Of 257 Sukur households in plains wards, 29 (11.3%) were ɗai, the rest mbəlim. These frequencies indicate that Sukur comprises more ɗai than most Mandara montagnard communities, a legacy of its famous former iron industry, and, although the difference is not statistically significant, there is an indication, supported by interview data, that proportionally more smith/potters than farmers have moved down to the plains since the 1950s, by which time smelting was becoming unprofitable. That relatively more smith/potters moved to the plains than farmers is not unexpected since cultivable land was available on the plains and ɗai had limited access to it on the mountain.

Table 1. Distribution of smith/potter (ɗai) households in Sukur Sama by clan and ward.














Gwafak -1; Daza – 1








In 1993 the 15 smith/potter households resident in upper Sukur resided in four wards (Table 1). Of these clans, two are long-established, the others relatively recent (see clan histories). The Təvwa clan is closely associated with that of the Dəmsa, who also reside in Gwassa ward, and their tradition has it that both clans are descended from sons of the same woman, one of whom showed a talent for the forge and the other for farming. Thus ɗai and mbəlim are brothers, as Mandara mountains myths often relate, and indeed these two clans have particularly close social relationships. The Təvwa households are all resident in the Ndilləi sub-ward of Gwassa, located on the northern paved way and close to the site of the former iron market (Fig. 2). The Dəmsa live in the neighboring Goeri sub-ward to the east.

Figure 2. The northern paved way (civi mungan) just below Ndillei (8 Dec. 1992).

The Təvwa are divided into two patrilineal sections, the people of the Tliɗi Ɗai (literally Chief of the specialists) and the people of the Ɗai Kərɓa (literally the Specialist of the stone slab) (Fig. 3). When two elders were questioned on the origin of the caste division in October 1992, they agreed that this goes back to the times before their supposed (and only recently claimed) migration to Sukur from Mpsakili (Gudur). A member of the Tliɗi Ɗai claims that they were the first to come to Sukur from Gudur; the Ɗai Kərɓa arrived later and were told by the Tliɗi Ɗai they could sacrifice to a nearby spirit for them.

Figure 3. The Ɗai Kərɓa, old and very frail, dances with women next to his slab during the Yawal Hidi festival (23 Feb. 1993).

The Rəvai, a long-established clan which includes both specialists and farmers, also claim a Gudur origin. After their arrival at Sukur one of them ate a squirrel and became a smith. Their two ɗai households, both resident in Duŋgom, include that of the Tlagama title holder and of Hundu, the most active and accomplished blacksmith in upper Sukur (Fig. 4). (There are scenes in and around Hundu’s forge in our video Black Hephaistos: exploring culture and science in African iron working (1995; 48 mins), freely available at

Figure 4. Hundu, a smith of the Rəvai clan, seen here during the Ɓər male initiation ceremony. Besides his work as a blacksmith he is a fine drummer (30 Aug. 1992).

Tlagama is the chief’s barber responsible for dressing the hairlock that marks his office and those of some neighbouring chiefs with close relations to Sukur. Tlagama is also one of the chief’s drummers and in the past had other responsibilities including the feeding of Hidi’s slaves awaiting sale (Fig. 5). A Rəvai farmer is one of the six Mbəzəfwai who at Zwaku make sacrifices at points on the various ways in to Sukur Sama to prevent the entry of evil.

Figure 5. Hidi, preceded by the junior Tləfu, approaches the northern entrance to the Buk enclosure during the Yama pə Patla ceremony while ɗai title-holders Tlagama (left) and Tləgəm beat theIr drums (28 Nov. 1992).

The other exclusively smith/potter groups are the Kwasha, Kwazhuwa and the Ka-Mariya. Both Kwasha and Kwazhuwa claim their ancestors originally came from Gudur. The Kwasha claim to have moved to Sukur from Dzu, a Margi community that apparently lacked smith/potters, and the Kwazhuwa clan from Futu Kakafu, a Higi-speaking community, where they were already divided into smiths and non-smiths. Other Kwazhuwa, non-smiths, settled in Margi villages and became Margi. The single Kwazhuwa household on the mountain is that of the Tləgəm title holder. Kwazhuwa are “ɗai to Shagwam”, while the Kwasha, resident in two Təka wards are “ɗai to Yanna clan”. This is to say that when the first members of these clans, attracted (we infer) by the employment opportunities offered by its iron industry, arrived in Sukur, the Shagwam and Yanna became their sponsors. To the best of our knowledge, while we were on the mountain neither Kwasha nor Kwazhuwa clan included an active blacksmith.

Unlike many other groups of the region, the Sakun members of the smith/potter caste monopolize only smithing and potting. Although burial ceremonies for important elders were once directed by the Təvwa clan Tliɗi Ɗai section, this is no longer the case and the title lapsed. Sukur smith/potters do not preferentially specialize in divination, healing or midwifery, although they include expert practitioners of all three and also drumming. It is unclear if there ever was a chief of the smiths who played a vital role in rituals held on behalf of society at large, although the Ɗai Kərɓa’s responsibility for overseeing the burial of the Hidi can be considered in this category. After an installation ceremony that parallels that of the Hidi, he cannot look again upon the chief in life. Other ceremonies in which he plays an active role are Hən dlə, Yama pə Patla and Yawal.

The traditional responsibilities of the Ɗai Kərɓa also include carrying out sacrifices for persons under attack from a certain kind of spirit, and allowing his house to serve as a sanctuary for someone who had committed a murder or serious crime. But unlike specialists in some other montagnard cultures, ɗai do not act at a familial level as priests in rituals on behalf of mbəlim. The relationship between smiths and chiefs is considered in greater detail by Sterner (2003: chapter 9).

Iron production
Until the 1950s, when imported metal and scrap rapidly replaced bloomery iron, Sukur was a producer and exporter of iron on an industrial scale (Sassoon 1964; Barkindo 1985a; David 1996; David & Sterner 1995, 1996; Smith & David 1995). Their furnaces could produce up to nine compact blooms in a day (Fig. 6). The iron was used for the manufacture of tools and a variety of weapons, ornaments and ritual objects, and dəɓəl, iron bars were exchanged in marriage payments and in trade with neighbours and northern groups such as the Kanuri of Borno.

Figure 6. Remains of a Sukur furnace from Gwassa ward and a sketched reconstruction showing the tall shield above the shaft and behind which the bellows were pumped, delivering air through a sub-vertical tuyère to the reduction zone at its tip (17 Jan. 1993).

The quantity of iron exported was very large, sufficient in normal years to manufacture over 50,000 hoes (according to N. David’s would-be conservative estimate). Unlike among the Mafa, Sirak and Kapsiki where only specialists smelted, virtually every family was engaged in iron smelting, conducted as a family business that might involve two closely related or sometimes close friends’ families working two furnaces simultaneously. Towards the end of the preceding rainy season ɗai women travelled up to twelve kilometres to collect magnetite ore, many having special friends in Kamwe villages. In the early dry season their menfolk cut and burned selected trees over an even larger area to make charcoal. Hidi negotiated with neighbouring chiefs to facilitate these arrangements.

During the March to May smelting season, furnaces were in constant use. Women not only brought drinks and food for the smelters and water to douse the blooms, but on occasion worked the bellows. This was inconceivable where smelting was restricted to smiths alone and is indicative of the at least partial secularization of the craft. Smiths and their families often smelted themselves, but their main work, carried out throughout the year, was to fine the blooms, i.e., to separate the slag and charcoal from the iron, and forge the wrought iron into the widely traded currency bars (Fig. 7). Sukur’s intensive iron smelting may account for the greater proportion of smith/potters in the Sukur population as a whole, 9.4% (according to our census data) as against the usual Mandara range of 2.5% to 5%.

Figure 7. A Sukur bloom about 18 cm in diameter and small currency bars of about 270 g. mean weight (28 Sept. 1992).

In the 1990s the Rəvai smith Hundu of Duŋgom was the only regularly active smith in Sukur-Sama. One Təvwa smith had a properly appointed forge, but worked only part-time. No other ɗai had such forges. Upper Sukur must have been obtaining a good proportion of its iron tools, weapons and other items from smiths on the plains at the Mefir Suku and other markets.

Funeral direction
Ɗai Kərɓa remains responsible for supervising the burial of Hidi but has no other funerary duties. Until the last Tliɗi Ɗai died in about 1985 (and was not replaced), he directed funerals of certain elders and played a special role at the Zwaku ceremony of purification.  Sukur smiths have not, so far as we know, conducted funerals for mbəlim since some time between 1984 and 1991. According to James Wade’s fieldnotes of 1984 the Tliɗi Ɗai was the sole individual responsible for burials. This is, almost certainly, a misunderstanding. In 1992 N. David was told that the Tliɗi Ɗai supervised smiths of his section of the Təvwa at burials of important elders. Their duties included dressing the corpse, carrying it on their shoulders to the grave and placing it in the grave. An elderly ɗai of Ka-mariya confirmed that Tliɗi Ɗai’s people were the only smiths involved in such elaborate funerals. A smith’s explanation of the change in funeral direction is that their participation was no longer worthwhile as the traditional payment of a goat plus a full basket of threshed millet had been reduced to the latter only. Others told us that the smiths refused to bury, saying that they “no longer eat useless things”. Amongst both ɗai and mbəlim, sisters’ sons (kənəi) and sons-in-law of the deceased now carry out at Sukur the duties performed elsewhere by casted smiths. These may include preparation of the corpse, digging of the grave, dancing with the corpse, transportation of the body and its placement in the tomb (Fig. 8). A sister’s son also instructs the inheritor in ritual matters.

Figure 8. A Sukur funeral of an old woman conducted by her brother's kin. In the middle of the picture her body, dressed in more or less as a man to honor the mother of twins, is danced on a man's shoulders. Meanwhile in the center foreground daughters-in-law wear traditional costume (17 Feb, 1993).

Many instruments including flutes and certtain kinds of drums are played by mbəlim, some only at certain times of year. As noted above, two ɗai titleholders, Tlagama and Tləgəm, are drummers. Smiths of these and other smith kin groups are hired to play at funerals and other public ceremonies (Fig. 9). In addition to Sukur smith-drummers, other musicians may be hired from nearby Wula settlements and from the more distant Higi community of Kamale to play at funerals and occasionally other ceremonies.

Figure 9. Ɗai drummers at the Ɓer male initiation ceremony (30 Aug. 1992).

We were surprised to find at Sukur not only young specialist women but some in their 40s and 50s who had never potted. This is despite a demand for pots such that practising potters are able to sell their pots almost as soon as they remove them from the firing pit. Much of the demand now and in the iron-producing past is satisfied by imports mainly from Higi and Margi potters. However, Sakun potters or resident immigrants make the pots for ritual purposes.

Ceramic vessels are at Sukur much less a focus of elaboration than amongst the Mafa and several other montagnard groups. Unlike their sister potters among the Kapsiki, Sirak, Mafa, and others, Sukur potters make few ritual vessels. Those that are produced are smaller and less carefully made, and potters are not importantly involved in any of the offerings associated with these pots. The past involvement of smith/potter women in smelting and in fining and forging bloomery iron, and their present involvement in farming have certainly contributed to the smaller numbers of active Sukur potters when compared with casted potters elsewhere, and probably also to the lesser elaboration of ceramics in Sukur culture.

Figure 10. A specialist makes a pair of pots celebrating the birth of twins. Note the anvil mould on which the pot was originally formed and is now supported. Her tamper is not visible (4 Dec. 1992).

Nonetheless, as in some other montagnard societies, Sukur potters have a ‘chief of potters’ (zəray ndeyi = woman + big) who is responsible for locating a clay source, making the first offering to it and advising others to make their offerings. No offerings are made to the potter’s anvil mould or ceramic tampers. The anvil mould might be made from a tree cut down by someone else and brought to a potter who shapes it herself. The firing pit requires offerings each year. This consists of a head of eleusine, a piece of bull’s hide, tobacco, germinated guinea corn, and sesame (the same as for a clay source). A small branch of white acacia (Acacia albida) is placed at the edge of the firing pit to protect against a sorcerer’s “evil eye”. Firing takes place at night when winds are weaker and for fear of the evil eye. As at Sirak a dog turd, suggesting to evil spirits that what is in the firing pit is completely useless, protects the pots from breakage in the firing pit.

At Sukur the midwife may or may not be a specialist. The most esteemed midwife in 1996 was a male ɗai not practising as a blacksmith who dealt primarily with difficult births. The female midwives are always old women, and indeed the word for old woman (zərajik) also means midwife. Some ɗai women specialize in treating children, and at least one magically removes foreign objects from the body. The male midwife mentioned above was also Sukur’s most famous diviner. Clients came to him from all over the region. He said he never farmed because between midwifery, divination and healing he did not need to: his wife and brother did the farming.

Figure 11. Gambo, midwife and diviner, receives the anthropologists and ND's daughter Branwen. Inset: crab divination, the various objects in thep ot represent people and things relevant to the divination (9 Oct, 1992).

The changing status of ɗai

Sukur smiths differ from their counterparts elsewhere in that neither are they exclusively ‘outsiders’, nor were they and the farmers all brothers (females are not noticeably present in these legends) in the remote past. Smiths play a role that is historic rather than mythical in the development of Sukur chiefly dynasties. Sukur specialists are categorized by mbəlim either as “people who do not eat bad things” or “people who eat bad things”. The people of Tliɗi Ɗai were classed as “dirty”, while those of Ɗai Kərɓa are regarded as much less so. The smiths of their neighbours are classified in the same manner, smiths of Roumzou, Margi, and Roumshi being dirty because of the foods they are reputed to eat. The Sukur state that their smiths are not like those of the Kapsiki or Margi, whose “eyes are like pepper and their children … ugly”. Sukur smiths “don’t eat useless things”. That they no longer bury any dead besides their own was never mentioned as a reason for their lesser degree of pollution.

Meek (1931a) was the first to mention the Sukur tradition (possibly now lapsed) that Hidis marry at least one woman of the smith/potter caste. No other farmer could do so, not, so we were told, because of pollution but because he would appear to be competing with the chief, an offence punishable by confiscation of property, exile or death. Some smiths agree that this was true in the past, but state that as long as 60 years ago ɗai men had married mbəlim women. We documented cases of smith women married to farmers, and met one smith on the plains who had been married to a woman farmer, all after the end of smelting. According to some, when a smith woman marries a non-smith she must first bathe standing on an anthill. James Wade (pers. comm.) notes that among the Fali of Jilbu the children of smith women who married farmer men are called grandchildren of smiths, not smiths. They are on the way to becoming ordinary people.

There has been a tendency over the years for smiths and potters to move to the plains. There is also an apparent increase in marriages of smith women to non-smiths, and for this reason and personal preference fewer ɗai women are potting. Since the Tliɗi Ɗai’s people no longer deal with the dead, they are more and more considered ‘good’. Is the distinction between the specialists and farmers breaking down? In 1996, shortly before leaving Sukur, N. David took shelter from the rain at a house with a simple forge in Fa, a subward of Dzuvok, located a kilometer east of the largest subward.

The presence of the forge was puzzling because this was the house of a farmer. He had set up the forge in 1993, but had already been smithing just outside his house for some seven years. People were shocked at first and told him they thought he had become ɗai. Now they joke about it. He acquired some of his tools from a smith who quit smithing (the latter was forced to farm full-time because he has many children and no wife). The new ‘smith’ is a Christian. Thus It appears that, over the past decades, Sukur caste boundaries have become less distinct and may eventually disappear.
Further reading
There is a substantial literature on the specialists of the Mandara mountains. Judy Sterner’s (2003: chapter 8)  “Smiths and potters: men of death – women of life?” surveys important aspects of their lives, providing many references to earlier work. Much information on Sukur presented above is taken from that fully referenced and footnoted chapter. More recent contributions include the edited volume Metals in Mandara Mountains society and culture (N. David, ed. 2012) and W.E.A. van Beek’s (2012) The dancing dead: ritual and religion among the Kapsiki/Higi of north Cameroon and northeastern Nigeria, Oxford: OUP, and his (2015) The forge and the funeral: the smith in Kapsiki/Higi culture, East Lansing: Michigan UP.

Culture Of The Mandara Mountains.