Pula: Botswana Joumal of African Studies vol. 15 no. 1 An ethnographic study of initiation schools among the Bakgatla ba ga Kgafela at Mochudi (1874-1988) Morongwa Nancy Mosothwane In and around the Makgadikgadi Pans there are several archaeological sites that have been interpreted as places were initiation rites took place in ancient times. But how can one tell they were initiation sites? This paper attempts to answer this by looking at modem initiation ceremonies among the Bakgatla of Mochudi and the "sites" they produce. Identifying ancient initiation sites ~ not be as simple as it seems. The village of Mochudi was established around 1871 under the leadership of Kgosi Kgamanyane. The decision to move to Mochudi was prompted by hostile relations between Kgosi Kgamanyane and Paul Kruger, one of the settler authorities of the Rustenburg District Kgosi Kgamanyane had at one point failed to provide enough labour to work in the Boer farms and in "disciplining" him, Paul Kruger had him flogged in public (Schapera, ~942). In 1870 Kgosi Kgamanyane and his followers emigrated from Soulspoort and settled at Tshwene- Tshwene for over a year before setting up the village of Mochudi. Nonetheless a substantial number of his people remained at Soulspoort and continued to recognize Kgosi Kgamanyane as their chief. Today Kgosi Lincwe II, Kgamanyane's descendant in the fourth generation, is recognized as the paramount chief of the Bakgatla 00 ga Kgafela living at Mochudi and at Moruleng. The purpose of this paper is to document the procedures of initiation ceremonies and to describe any physical evidence of initiation on the landscape and elsewhere. Other scholars have documented archaeological sites associated with initiation ceremonies of various Bantu communities, for instance T.N. Huffman and 1. Blacking on the Venda of South Africa (Blacking, 1969; Huffman, 1996) and A Richards on the Bemba of Zambia, (Richards, 1956). Huffman and Blacking have documented domba sites (domba is a Venda word for initiation) explaining how they may have been made, their architectural styles and other aspects such as material culture (like figurines of clay or wood), songs and dances. Likewise Richards has written extensively about the Bemba's chisungu initiation ceremonies (Richards, 1956; Huffman, 1996). Initiation is a common rite of passage among many world communities. It is usually associated with initiates' introduction to adulthood (Blacking, 1969). However, initiation schools among the Bakgatla ba ga Kgafela were not necessarily associated with physical maturity: very young people and mature adults were often initiated together. This kind of age distribution was partly a result of the criteria used for identifying people who were to be initiated together, and partly due to the absence of people from the village at the time of initiation (van der Vliet, 1974; Grant, 1984). For instance, the 1982 initiation of Mangana was led by a ten year old but a large number of initiates were over 20 years (Grant, 1984). In 1985 Bakgatla Kgafela led the Manoga at the age of 10 and his younger brother, Mmusi Kgafela, led the Matukwi in 1988 at the age of 9. Within the Matukwi there was Thari Pilane aged 6 years. All these mephato (a group of people who have been to the same bogwera or bojale schools) had youths over 18 years of age. Most of the detailed studies of initiation have been on male initiation schools, the few ~ deal with female initiation are those that are concerned with female genital mutilations as It has been observed among many East African societies. This study gives an account of both male and female initiation ceremonies. The study attempts to present a detailed record 144 of how people looked, danced and behaved. It gives details of initiation costumes, material culture and also ~e material evidence left on the landscape that will eventually become part of the archaeolOgical record. Girls' initiation schools are known as boja/e / bya/e while ?<,ys' schools are bogwera. Initiation has been an important component of Kgatla customs, It was ~ means of o~de~g the society, a lot of emphasis being put on the importance of respectmg ~lders as Identified by the order of mephato. Mephato were an important source of commUlllty labour and male mephato were used in times of social upheaval within the community to restore order and also in time of clashes with other tribes (Schapera,1942). The popularity and importance of bogwera and boja/e have gone through periods of decline and resurgence. It declined mainly due to lack of interest on the side of dikgosi who during their reigtls were not inclined towards promoting initiation. Except for Kgosi Kgamanyane, Kgosi Isang and Kgosi Lincwe II, other dikgosi who have ruled at Mochudi did not show much enthusiasm on the issue of initiation. The popularity of initiation was further weakened by the government and churches (Grant, 1984). These two factors together with the fact that initiation was kept a secret have led to the loss of important information regarding the way in which initiation was conducted in the pre-European period. Every time the institution was revived, it was modified to meet the recommendations set by both the government and the church. At the time of research there was a degree of secrecy and reluctance to talk about bogwera and boja/e in the community. I consulted Kgosi Lincwe II and his wife, Mrs. Mmarakau "Mma Seigwaeng" Kgafela who then introduced me to some village elders whom I interviewed. I was then able to identify some people who have played important roles in bogwera and boja/e. For the most part, individuals were interviewed separately but in cases where people were found in groups, they were interviewed at the same time. Literature available on the topic of bogwera and boja/e was also evaluated. Literature review Initiation rituals of different Bantu societies have been docun1ented by researchers from different fields of study such as history, anthropology, ethnography and archaeology. Written docU111ents appeared mostly in the early 19th century when more Bantu groups came into direct contact with Europeans. However, no attempts have been made to merge the historical and ethnographic records of Tswana initiation schools and their implications for archaeologists. A multifaceted research of this kind has been done on the Shona and the Venda (Huffman, 19%) and it was this research that became the basis for my project. Initiation ceremonies of most Tswana speaking communities follows the general pattern of initiating people in large numbers, male circumcision, formation of mephato and sometimes female genital operations which were equated to male circumcision (Willoughby, 1909; Roberts & Winter, 1915; Schapera, 1955; Monnig, 1%7). The mode of teaching was songs associated with moral referents and secret meanings. Some of the songs were not fully understood by initiates until they had becom~ ini~~on ~~rs .themselves (Schapera, 1978). Initiation songs and dances were the roam actIVity m wbatlon schools (Willoughby, 1909; Schapera, 1955; Richards, 1956; Monni~ ~967; Blackin~. ~~9; Huffman, 1996). Singing was the major mode of commuwcauon between wuat10n participants. Traditional musical instruments were used. The instrument that was used by women of different communities is the drum. For instance, it was used by the Bemba (Richards, 1956), the Venda (Blacking, 1%9; Huffman, 1996) the Pedi (Mo~g, 1967) and the Bakgatla. In all communities mentioned, the women's drum was trea~ With respect. as it was laden with ritual sigtlificance and symbolism. It was the central object that earned powers that ensured the smooth running of the ceremonies. It was therefore played only by 145 a few selected individuals who were known for their wisdom in initiation rituals (Monnig, 1967) When not in use it was often kept safely at the royal residence. The most common musical instruments among men in initiation schools were animal horns (Willoughby, 1909; Roberts & Winter, 1915; Blacking, 1%9; Schapera, 1978; Grant, 1984; Huffman, 1996). Animal horns appear to have been used for practical reasons only, any participant who knew how to blow an animal horn could do it during men's initiation ceremonies. At the end of the ceremonies those horns remained private property of individuals. There was no limit as to the number of animal horns that were used (Willoughby, 1909; Blacking, 1%9; Schapera, 1978). Among the Tswana, there was a tendency to put girls in seclusion in the village (Willoughby, 1909; Schapera, 1955) while boys were taken away from the village during initiation ceremonies. Boys were put under absolute separation from village life while girls were put through a period of limited social contacts only since they stayed in the village. This contrast on the use of space is an important one for archaeologists because it helps to explain the absence or presence of material culture and features associated with initiation ceremonies. A good example of an archaeological research on initiation materials and features and their relation to the use of space is by Huffman (19%) among the Venda. By mapping out a traditional Venda village, Huffman was able to identify house remains that were used for female initiation ceremonies. He also identified stone cairn fields as archaeological sites of male initiation. Hearth remains encircled by stones have been documented among the Pedi (Roberts & Winter, 1915) and among the Bakgatla (Willoughby, 1909; Schapera, 1978) but whether or not these are to be equated with what has been mentioned in Huffman's text (e.g., at Mosu and Kubu) is a research question on its own. Tswana communities initiate youths in large numbers to form mephato. Reasons for initiating people in large numbers have not been investigated fully among the Bakgatla but it is possible that it was meant to emphasize tribal unity. Initiates were taught the importance of unity and co-operation with one another and with the rest of the community. Among the Nuer it is thought that if a boy is initiated alone he would be lonely and he would probably die out of loneliness (Evans-Pritchard, 1940). Tswana initiation ceremonies led to the formation of mephato which grouped people of the same generation, although there were usually a few exceptions. Before western education, people had no means of telling their absolute ages and therefore used relative ages based on the order of mephato (Willoughby, 1909). In social gatherings and in tribal meetings, seniority of peoples was observed in terms of their mephato and that was probably why uninitiated men (magwane) were not allowed to participate in social decision making (Willoughby, 1909; Breutz, 1953; Schapera, 1978). Many anthropologists describe initiation schools in terms of education, where initiates are taken into custody of their seniors who teach them tribal laws, fighting skills, hunting skills, initiation songs, tribal folklore, craft making and many other things. One of the main objectives of initiation ceremonies is to train initiates to have courage and endurance (Blacking, 1%9). They are treated harshly, made to perform dubious tasks and are often beaten on the slightest pretext. However, the ultimate goal of initiation is not so much to make a stronger or even more knowledgeable person of the initiate, even though this may be greatly desired, but it is to transform the initiate, making him different from what he bad been. and separa~g him from his childhood existence (Lincoln, 1981). With the commg of western civilization, a lot of traditional institutions went through some changes (Blacking, 1%9). The role of initiation was being erased by the new forms of authority but this was a change not welcomed by traditionalists, as depicted by the 1920s 146 and 1930s relationship between the protectorate government and the "natives" . The protectorate ~overnment wanted to stamp out initiation schools because they considered them barbaric, cruel and unhygenic (BNA S217/6 & Sloo/2). Tswana traditionalists ~a~~u~ their good old tradition, while the government called for a replacement of IDloatIon schools by western education systems. Gradually, important Tswana leaders be~ to agree with the protectorate government and among them was Kgosi Lentswe I, KgOSlBathoeng II, Kgama III and Kgosi Sebele (BNA S10012; Schapera, 1936, 1978). On ~~ ~~ber 1917, the High Commissioner passed a proclamation that was to prohibit IDltIatlOnm the Bechuanaland Protectorate. No persons under the age of sixteen were to be initi~ted without the consent of their parents or guardians and anyone found guilty of forcmg young people to undergo initiation was to be fined not more than fifty Pounds or was to be imprisoned for not more than a year. The proclamation was initially meant for some parts of the Protectorate but the Resident Commissioner had the power to expand its area of application. The proclamation had not included farms owned by whites and therefore people around Lobatsi went and performed their ceremony in a farm. In Mochudi it was also continued although circumcision had been stopped. Between 1944 and 1945, the High Commissioner sent despatches to different Resident Commissioners asking whether the 1917 proclamation was being obeyed or not. Kgama, Sebele, Lentswe and Bathoeng were praised for having been loyal to the laws of the High Commissioner. Mephato of Bakgatla 1874-1985 Mochudi was first occupied by the emigrated group of the Bakgatla under the leadership of Kgosi Kgarnanyane Pilane around 1871. He had earlier settled at Soulspoort where his bad relations with the Afrikaner government eventually led him to leave the country to go and settle at the present day capital of the Bakgatla ba ga Kgafela, Mochudi, having stayed at Tshwene Tshwene first for a year (Schapera, 1942; Breutz, 1953). It is not clear as to which mephato was initiated first at Mochudi. Documents available indicate that the Matlakana were initiated in 1874 but before them was the Makgalo, a women's mephato whose date of initiation is not yet known (Breutz, 1953; Schapera, 1978). Evidence from Kgatla tribal laws tends to suggest that the Makgalo were the first mephato at Mochudi on the basis of the Kgatla tradition that in a newly established village the first mephato must be that of women. However, it is difficult to conclude that Makgato were indeed the first in Mochudi because before finally settling at Mochudi, the Bakgatla stayed at Tshwene-Tshwene for over a year where Makgalo may have been initiated. But given the circumstances under which Tshwene-Tshwene was occupied (a temporary village while land was being negotiated with Kwena authorities), it seems more likely that they were initiated at Mochudi. Kgarnanyane died in 1874, having been a leader of the Masoswe of 1849, and was succeeded by Lentswe, a leader of Matlakana of 1874. During his reign, Lentswe named eight mephato, Maisakoma, Makoba, Majanko, Matshabatau, Majekere, Mantwane, Mabusapelo and Makuka. In 1892 he was converted to Christianity and having adopted this new religion he was compelled by Christian ethics to prohibit some Kgatla customs such as paying bride price, polygamy and circumcision (Schapera, 1936, ~~42; BNA S2!7/6 & S218/1O). The Dutch Reformed Church (ORC) that Lentswe 1 had Jomed was partlcul~! against circumcision in bogwera schools and therefore ur~~. ~ntswe I to prohi~lt initiation. He informed his people about his decision t~ stop l~~tI~n but was met With criticism and refusal. The Bakgatla made it clear to him that ImOatlOnwas a communal entity, that no single individual could make a decisi?n to ~op it without the conse~ .o~the whole merafr. Lentswe I therefore lost his case against his people and eventually ImtIated 147 the Mabusapelo in 1895. According to Mr. Moreri Tshite the name was derived from the phrase 'go busapelo', which means to calm down. Thus, by selecting this ~ic~. ~e Lentswe I was saying that he had withdrawn from his intention of cancelmg mItIatlon. Having initiated this mephato, Lentswe was put under even more pressure by the church and since he was spending much of his time at the church he agreed once again that there would be no more initiation schools. On the other hand a large group of magwane was preparing itself to undergo the ceremony, pressuring Lentswe to authorize their ceremony: In 1902 he did finally let the youths be initiated and he called them Makuka. Mr. Moren Tshite has said that the name was derived from the phrase "go kuka", meaning "to lift", which implied that Lentswe I was never going to allow initiation schools with circumcision rites and indeed the Makuka remained the last mephato to have undergone circumcision rites for over seven decades until Lincwe II reintroduced it in 1975. Machechele and Mafatshwana were also initiated by Lentswe I but by then circumcision was completely prohibited. Lentswe I died in 1924 having retired earlier because of ill health (Schapera, 1942). lsang took over as a regent from 1920 to 1929 when his nephew, Molefi Kgafela, was crowned. lsang was instrumental in reviving the tribal customs that Lentswe I had abolished; he encouraged people to pay bride price, he recognized the traditional purification of bereaved people (baswagadi, batlholagadi and matlhomela), reintroduced rainmaking rituals and advocated for better initiation schools in which boys stayed away in the bush for months learning traditional songs and tribal customs, but did not reintroduce circumcision. Initiation gained more popularity unlike during the last years of Lentswe's reign where people were called to the kgotla to be given a mephato name without having learnt initiation songs or camping in the bush (Schapera, 1942). lsang's mephato were Maatlametla, Madima and Machama. In 1929 Molefi, the rightful heir, was crowned chief, taking over from his uncle lsang. He initiated Mafiri and Mangope. Molefi' s unacceptable behaviour and misconduct put him at loggerheads with the protectorate administration so much so that in 1937 he was forced to leave the Kgatla Reserve (Schapera, 1942, 1956; Grant, 1984). His younger brother, Mmusi, was then appointed as a regent. Mmusi initiated Malwelakgosi and Mathulwa and later on in 1963 Lincwe II was crowned. Mmusi was not very much interested in initiation and during his reign the custom lost its popularity and much of its significance. Boys were no longer staying a long time in the bush as lsang had insisted. For instance in 1963 the Mathulwa who were led by Kgosi Lincwe II spent only o?e night on the outskirts of Mochudi and returned early the following day to be named and dispersed. The initiation of Mathulwa was meant to formalize Kgosi Lincwe II's new posi~.on in a manner that was supposed to be traditional and yet a far cry from the traditIonal Kgatla culture. Traditionally, Lincwe II ought to have been initiated before being crowned but in his case he was crowned on April 6, 1963 and initiated later in June 1963. On the Monday following his initiation he was due to attend a court case in Molepolole where there were disputes over chiefs and as a result his initiation ceremony had to be as short as possible. For ten years after the initiation of Mathulwa no new mephato were formed and it was assumed that initiation had been abandoned. Shortly after his installation Kgosi Lincwe II left the country to go and work in the United States of America while Mmusi acted on his ~half at Mochudi. Before he left he initiated a women's mephato of Maisakoma and upon his return a male mephato was formed. This was Masoswe and for the first time since 1902 ~~cision was performed. Kgosi Lincwe II did not insist or force people to go for lDltIatlon but once enrolled there was no way of escaping circumcision. Circumcision itself was done at the Deborah Retief Hospital run by the ORC; the very same church that bad 148 ~~i.s«:d on the abolition of circumcision earlier (Grant & Kgafela, 1989). In the later lllitlat~on.ceremonies students from the Medical University of South Africa (MEDUNSA) were mVlted to help because of the hospital's staff shortage (Grant, 1984). Incorporation into mephato !here are several ways of identifying people who are to be initiated together. Membership mto a mephato is not by an individual's choice but the system has criteria that are used to select people to be initiated together. The following are the criteria used. Eldest sons of men of the same mephato are initiated together so that the younger mephato is identified as a "child" of the older one. For instance, the Mathulwa are sons of Machama. Mr. Segwati Ramontie was the eldest in his family and his parents were of Mantwane and he became a member of Madima. Madima are identified as sons of Mantwane (see appendix for more details). Once eldest sons have been initiated their younger siblings may then follow in their order of age. 'No two brothers or two sisters from the same mother and father are allowed into the same mephato. Unless they are twins, they are only allowed if they have only one parent in common as is the case with Mr. Ray Molomo and his brother Mr. Norman Molomo who are both members of Mafiri: they are of the same father but have different mothers. Informant say that a brother and sister, of the same parents, following each in their order of birth may share the same mephato name. Those people who may have been absent when their turn for initiation was due were initiated with younger mephato but given their proper mephato names. The general sequence of mephato is that of men followed by women but there are occasions when only women are initiated without a male counterpart. Those female mephato without men are; Mantshakgosi, Makgalo, Maisakoma, Matshabatau, Majekere, Mabusapelo, Maatlametla, Matshego, Malwelakgosi, Maisakoma and Mabusapelo. In cases where the village has been relocated the first mephato in the new village was that of women, for instance Makgalo may have been the first mephato at Mochudi initiated between 1869 (Mafatlha) and 1874 (Matlakana) with Mochudi having been occupied in 1874 (Schapera, 1942; Breutz, 1953). A village was equated to a household and a woman was the one responsible for the daily chores needed to maintain the family. A woman therefore played an important role in the house and it was eventually reflected in the society. Every time a new chief was installed the first mephato should be female and hence the saying that "Kgosi 0 itibola ka basadi" that is a chiefs eldest "child" is a woman. The idea is to show appreciation to the chief s mother who has over the years taken care of her baby, a baby that had grown strong enough to take care of his father's people. Mephato formed for this reason include the older Maisakoma (initiated by Lentswe I), Maatlametla named by lsang, Matshego named by Molefi and the younger Maisakoma named by Lincwe II. Some female mephato were formed probably because of higher female population and probably because more girls were needed as wives in a polygamous society that insisted on initiation before marriage. Bojak / byale: female initiation practices Bojale is a female initiation ceremony through which girls are incorporated into mephato. Once the boys had been sent out to an initiation school, girls automatically knew that they would be sent for their initiation ceremony the following winter season. Girls do not have to undergo the laborious pre-initiation sessions that the boys undertook (bogwane). Bojale like other components of Kgatla culture has been changing over the years. There has been some swift changes in the manner in which bojale is carried out especially from around the 149 colonial period when a new religion and government were replacing the traditional religious beliefs and traditional authorities. From around the late nineteenth century up to the early twentieth century there was an initiation style that was abandoned probably after the initiation of Makuka in 1902. Only a few informants had vague recollections of what they had heard about that style of bojale. It is said that girls were initiated within compounds of carefully selected families, mo~y those related to the royal family. Tightly packed branches were used to seclude the grr~s away from the rest of the community. They stayed in total seclusion throughout therr initiation periods and initiation was kept a profound secret from non-initiates; even those who had undergone the ceremony did not talk about it openly. Only a few informants were able to remember this long forgotten tradition. It is not known when and why initiating girls within compounds stopped. Perhaps it was due to population increase, when it was no longer easy to accommodate the initiates and their teachers in a regular sized compound From around the 1920s a different style of initiation was introduced and it was mainly due to criticism of the custom by the Dutch Reformed Church that was by then fully established in the village. Many important components of bojale appear to have been abandoned in order to suit the expectations of the church. Girls were no longer initiated in compounds and they no longer spent the whole winter season in an initiation school. They were instead called to the village's main Kgotla where they were given a very light traditional whipping (go loma tsetse) and then a mephato name. Initiation was no longer of any great importance in the society and incorporation into mephato was no longer a source of pride (Grant & Kgafela, 1989). Ultimately in the mid twentieth century, initiation regained a higher level of popularity because the new chief, Isang, had put more emphasis on the importance of mephato as a source of community labour and organization. Instead of a one day initiation ceremony, girls now stayed up to six weeks learning initiation songs, tribal folklore, social and moral behaviour and the responsibilities of wives and mothe~ (Schapera, 1942). This remained the standard procedure of girls initiation ceremonies unul the late 1980s, except that the duration of the ceremony has been decreasing slowly because most initiates were either students or workers. Every late afternoon initiates from each sub ward (kgoro) met at their meeting places and when all had come they went to their ward's central place where all initiates and their teachers from the ward met. While waiting for other initiates to arrive the girls sang under instructions of their teachers. Initiates from the different wards would then march to the main kgotla where they received instructions of which routes they would take that day. They then made a procession, singing all the time and occasionally stopping for a short rest. They were sometimes instructed to jog or jump over small thorny bushes and all the time they had to keep their heads bowed Initiates were not allowed to walk up straight and even wh~n resting they either had to kneel down or sit in a squatting position without spreading therr legs. Those postures were a sign of respect for the teachers. The initiates bad to keep their arms folded on their chests because they did not wear anything on the upper body (Blacking, 1%9). ~ere were. two sessions held every night. The first session began just after sunset and contInued until shortly after midnight while the second session began in the early morning hours and lasted until just before sunrise. The reason for doing so was that students and workers could ~hoose sessions they felt comfortable with. Only those people not going to school or working were expected to turn up for the two sessions every night for they bad time to s~eep during ~e day. I~ was only on the last day before graduation that there was one all mght long session and m the morning the initiates were laid down in their order of seniority and whipped. The whipping was known as go loma tsetse or "to bite a tsetse fly". 150 Initiate~ were often taken to open grounds within the village where they danced and sang. ~~se mcluded Seboeng, Sethobong and Phaphane show grounds. From around the 1980s ~tlates went out as far as Mmaesoke, a ploughing area about 5 kIn to the northwest of the ~llage. Mmaesoke was chos~m as ~ ini~ation ground for girls because the grounds in the VIllage were now developed mto reSIdential areas, while in other cases bars were built next to such grounds. ~a~soke was .not very far from the village and was large enough to ac~omm~te all partlClpantS. During the day, participants stayed at home and went around domg therr day to day chores. They were not in anyway distinguished from the rest of the people for they dressed ordinarily, ate everyday food and were not put in seclusion. Unlike ~ys. who p~ctically moved out of the village during their initiation periods, girls remained Within the VIllage and continued to live their ordinary life during the day. Girls' activities began at night and ended before sunrise. The traditional attire of initiates was makgabe and mothikga. This attire was used for a very ~ong time until it was ultimately replaced by European cloth at the beginning of the twentleth century. Makgabe is a kind of skirt that is made from separate threads sewn together at one end and hanging loose at the other end It was made from a bush plant called bogokgwe or mosoke/atsebeng whose roots and stem were harvested from swampy areas. The plant was boiled for several hours to remove its green pigmentation and to soften it up. It was boiled until it became clear coloured and tender. Threads that carne after boiling were then rolled between the thigh and hand Pieces of roots and stem were processed into one long roll of thread which was then cut up into many pieces that measured from the waist to the knee, and the pieces were sewn together using very thin thread obtained from an animal skin. Charcoal mixed with animal fat was then smeared on makgabe and rubbed until it had a shiny black colour. Small knots were then tied up on each thread as a way of decoration. Since makgabe did not cover the buttocks, an animal skin back apron was made for this purpose. Goat or sheep skins were softened and cut to knee length with belt extensions on either side of the waist for tying up the apron. Except for these two pieces of clothing, initiates remained bare (Blacking, 1969). Women in the village and presumably initiation teachers wore leather skirts called mothikga made from goat or calf skins. In hot weather women remained topless but in cold weather they wore small leather blankets which were tied around the neck like capes (Blacking, 1969). No forms of body decoration were used on Bakgatla girls during initiations. If one had any item of decoration such as necklace or bracelets she had to take them off when going for initiation sessions. One of the most common ways of decorating the body was body painting done by other Tswana communities like Bakwena and Balete (Schapera, 1955; van der Vliet, 1974). These communities used red ochre to paint either initiates or brides. The Bakgatla believed that a woman was to be born out of a girl initiate and the newly born woman had to resemble a newly born baby as much as possible and therefore initiates were stripped of all decoration items. However, Mrs Fenenke Mmutle mentioned that her mother had a set of copper bracelets that was part of her regiment's regalia. The bracelets have long been lost and therefore attempts to have them drawn or photographed failed None of the informants nor the Phuthadikobo Museum in Mochudi had any documentation of such bracelets. There is not much material culture that the Bakgatla women used or produced during their initiation ceremonies. Many southern African societies are known to use a variety of figurines in their initiation ceremonies. People like the Venda (Blacking, 1969; Huffman, 1996) and Bemba (Richards, 1956) relied heavily on the use of figurines as teaC~ aids in female initiation schools. The Venda even used figuratively carved drums during domba ceremonies (Blacking, 1969). During the chisungu ceremonies, the Zambian Bemba people 151 used a variation of special figurines and distinctive architectural enclosures where girls were kept in seclusion. The Bakgatla women did not use any figurines in their boj~le ceremonies. Their mode of teaching was songs. Bakgatla women even though lacking figurines had a drum that they used. The drum was used only by women in their initiation ceremonies. The drum was used to symbolize a womb with its bottom opening symbolizing the opening of a womb. The drum was long and narrow with only one carved handle. Its shape is not in anyway similar to that of a womb. The drum was never to be touched by men and when played in traditional events at the kgotla in the presence of men, its lower opening was never to be directed towards men. Not every girl or woman could play the drum: only those with special posts in initiation schools, such as the royal mother's close councilors who could make decisions and suggestions to the rest of the troop. One such person was Miss Mogaritshi Pheto whose mother had also been known for her wisdom in initiation procedures. It was believed that some people in the group had strange evil powers or were witches and if they were to play this drum disasters could befall the initiates. Even though there was a traditional cleansing or healing at the beginning of initiation ceremonies, it was always advisable to avoid taking chances with witchcraft and that was why playing the drum was reserved for only a few people in the village. At present the Phutadikobo Museum is housing the women's drum. The drum is probably the only artifact in the museum collection that the Bakgatla brought with them when they migrated from Pilansberg in 1871. Besides the drum, initiation teachers carried with them branches of moologa tree (Croton gratissimus). They enclosed initiates in the center and moved very closely to each other and the branches became tightly packed, completely screening away the initiates in the center. No reasons have been found as to why only moologa branches were used instead of ot~er tree branches and therefore the significance or symbolism of moologa branches rernams unknown. At the end of the ceremony the branches were discarded. On the last night of the ceremony when initiates stayed all night long for dikgalaopa, there was ritual whipping, go loma tsetse, of the initiates who were ordered to lie down on their stomachs and were beaten using moretlwa sticks (wild berry, Grewia flava) that had been previously prepared. Since initiates were large in numbers several sticks were used for this ritual because they would break. Like moologa branches the sticks used for whipping the girls were afterwards discarded. At the beginning of each initiation ceremony initiates were traditionally medicated (ba a phelwlwa). The idea was to guard them against evil ancestral spirits and at times of inter- tribal wars they were to be protected from possible witchcraft and curse from enemies (Blacking, 1%9). A young virgin was selected from a humble family and one that was related to the royal family. She was accompanied by the royal mother and her assistants to go out and collect plants used in the ritual. The women identified the plants which only the virgin was allowed to dig out or else they would become contaminated and loose their power if .dug out by a non-virgiD. The plants were then treated accordingly at the back yard of the chief s house to produce the traditional medicines needed. Plants used for go phelwla remained an absolute secret from the rest of the people so much so that even in the 1980s ~~:n there was less emphasis on keeping the procedures of bojale a secret from non- tnIbat~, .~es .of th~ plants used for medication continued to be a profound secret. None of the tnItmtes mtervtewed knew what plants were used, only Mogaritshi Pheto and Mrs Mmarakau Kgafela admitted that they knew the plants by names. They strongly believed that I wo~d become insane if they were to tell me the names of plants. A small fire was set ~t the ~ entrance to the kgotla and herbs used for medicating the initiates were thrown mto this fire, the drum was suspended above the fire to let the smoke flow inside. It was 152 believed that by so doing the initiates' wombs were being cleansed to be able to bear ~hildren. All initiates who had small babies were to squeeze some milk on to the fire to put It out and by so doing they were protecting their children. Other initiates then walked over the fire spot bare footed and the ritual would be completed. All initiates were warned a~st making and using their own personal traditional medications during the initiation penod or else communal medications would work badly against them. There is, in fact, a sto~ ~f one woman who during her niece's initiation came with her own package of medicmes and just as the initiates were entering the main kraal to be named the woman fell down and was paralyzed instantly. She is said to have died a few weeks later. Phek% was done only on the first day of the ceremony and the powers of the ritual were believed to be strong enough to last an initiate's life time. Many people believed that nothing bad would happen to them once they had been medicated. They said they feared no snakes and other night creatures when they were perfonning the ceremony at night. Some symbolic actions have been identified during boja/e ceremonies. The actions are associated witlI child bearing which is an important aspect of societal growth. The Bakgatla believe that a mature woman was born out of a girl initiate and therefore most of the songs sung and actions perfonned center around child bearing. In tlIe most commonly perfonned action initiates walked with their heads bowed, arms folded over breasts and sitting in a squatting position. Initiates were expected to walk with their heads bowed or else they would be beaten for failing to do so (Blacking, 1969). Since they were imitating a fetus which cannot look around and see its surrounding, the initiates had to be forbidden from looking around. They folded their arms in front to hide their breasts presumably because as young children they were expected not to have breasts. When stopping for a rest, they either knelt down or sat squatting and botlI positions are similar to that of a fetus and therefore the assumption is that these postures symbolized fetal postures. When initiates left home for their sessions they left fully dressed because it was nonnally just before sunset, with people moving around and about. As soon as darkness fell they were ordered to undress and they remained with panties only (Kollars, 1991; Motlotle, 1998). The clothes were carried by teachers, wrapped in blankets and carried like babies on the back. This manner of carrying clothes was meant to encourage initiates to get married and to have children. Male initiation A group of boys preparing tlIemselves for initiation was called magwane. Magwane stayed in groups doing daily chores at the cattle posts and fields in tlIe rainy season and in winter they idled around in tlIe village. They often sang intimidating songs and provoked people because in Kgatla custom they could not be taken to court (Roberts & Winter, 1915; Schapera, 1978). They misbehaved mostly at night when it was not easy to be identified or else tlIey would be ill treated when their time for initiation came. Periods of bogwane lasted as long as three years until the chief and his councilors felt that the boys were strong enough to undergo initiation. Magwane had a ritual of stick fighting where they divided themselves into two teams and fought each other using moret/wa sticks (Schapera, 1918). They had tlIeir own style of clotlIing that distinguished them from the rest of tlIe boys. They had head gear from wildebeest (kgokong). A strip of skin was cut from the ~ldebeest's spine and was tied into a circle to fit an individual's head When they left the vIllage to be initiated in the bush they left the head gears behind because tlIey were no longer s~ll ~ys as tlIe gear signified. Magwane also wore leather aprons known as diko/a ~g ~ck fights. Diko/a were normally made from calf skins but could also be decorated WIth stnpeS 153 of tshipa skins. From around the 1930s buttons were commonly used to decorate belts worn bymagwane. Once enrolled into an initiation school, magwane were now called magwera. Magwera had a different code of attire. They wore leather underwear known as phuduhudu or moswapo, and leather shawls known as mokob%. These were made from any ani~s killed during the ceremony. When returning from the bush they carried ostrich feathers tIed to sticks and these were known as mokob%. Initially all initiates were given mokob% but in the later years with an increased population and control of hunting by the Wildlife Department only senior boys from different wards were given mokob%. If the group was led by an heir to the throne, he would wear mokob% made from a lion skin because it was a fierce animal but less significant than leopard (the chief wore leopard skin). For instance in 1982 Kgafela Kgafela wore mokob% made from lion's skin. In the later years people began to wear short trousers and jerseys during initiation ceremonies. Mokob% were put over jerseys and rifles became increasingly common. Leather sandals were replaced by European shoes. Bogwane has long been abandoned and now madi tse a dikgokong are worn by initiated men to decorate themselves. While in the bush magwera spent their time hunting, singing and doing crafts. Hunting was important for meat and leather for making clothes. Besides meat, participants ate porridge regularly supplied from the village by ox-wagons travelling between camps and Mochudi. The most favorite food appears to have been /ebabe made by frying and grinding sorghum grains. After grinding, a powder of moret/wa was added to sweeten this sorghum powder. Lebabe was made by women in large sacks and was collected by initiation teachers. It did not go bad quickly, and once prepared it could be eaten with water without having to cook again. While in camps, initiates made wooden utensils like spoons, stools and leather strap chairs. Those interested in leather products engaged in softening and cutting leather from hunted animals. Initiates camped at different places in the bush and each mephato chose its own camping area. This was determined by the availability of water and wild animals. In some camps they stayed for a single night while in some they stayed for over a week depending on water sources available. Figure 1 shows some of the boreholes and streams in Kgatleng by which bogwera camps have been set in the past. Initiation sites There is variation in the potential archaeological record that successive male mephato produce. This is mainly dependent on the manner in which different chiefs viewed bogwera and the manner in which they ultimately changed or altered bogwera proceedings. A lot of changes occurred in the early twentieth century with influence from western civilization (Schapera, 1936). Each chief was influenced one way or the other and it was this influence that determined the direction of bogwera and boja/e proceedings. Evidence of initiation procedures left behind may be in the form of initiation sites and their associated features and material culture. A number of initiation sites have been identified in and around Mochudi. Tsope is a valley located about six kilometers west of the main kgot/a. It is an area associated with bogwera camps. It is a small valley area bounded by a series of small hills that provide a natural screen away from the village (Fig. 2). The site consists of a small stream that provided water.for people in bogwera camps. It was an ideal place for setting up a bogwera camp because It was not very far from the village, which made it an easy task for village ~omen to take food to the camp. It was naturally secluded from the village and it also had Its own water source. Tsope was established as an initiation site around 1874 and as far as oral traditions are concerned only five mephato were initiated at this site: Matlakana 154 ~1~74), ~oba (1880), Maj~o (1884), ~twane (1892) and Makuka (1902). Initiation signature.s at Tsope are mainly hearths encrrcled by stone slabs (Fig. 2). Slabs of stones were obtamed locally and planted vertically on the ground to make stone fire places. The first slab planted was smeared with traditional medicine made from foreskin residues of previous circumcisions (tshitlho) (Schapera, 1978). At the time of investigation a lot of these hearths had been totally destroyed as the site was developed into a residential area in the 19~0~. Unfortunately no attempts had been made previously to map Tsope and as a result It IS not easy to reconstruct the layout of the original site. At present only fifteen h~arths, of which four are hardly intact, remain in an area of about 160,000 sq m. Their diameters are mostly around 1 m. A layout of a bogwera camp is similar to that of the main village in terms of how other wards are located away from the kgosing ward (Willoughby, 1909; Grant, 1984) but attempts to locate where some wards may have been were fruitless because there was no means of identifying the kgosing hearths and also because it is not known exactly how much of the site has been destroyed. Molapo wa baruti or Kgogometsong is another circunIcision site located on a stream bed about one kilometer west of the village hospital. It was first used for bogwera purposes in 1975 by Masoswe who were initiated by Kgosi Lincwe II. It has since been used by Madingwana (1976), Malomakgomo (1980), Mangana (1982), Manoga (1985) and Matukwi (1988). Kgogometsong was chosen for bogwera camps because it is on a streanI bed bordered by a small escarpement that shields it from the rest of the village and most importantly because it is very close to the hospital which had taken over the responsibility of performing circumcision operations. Water was not so much an important factor as it could be fetched from nearby taps. Unlike at Tsope, hearths at Kgogometsong are not encircled by stone slabs but they just lie in the open Ash deposits of this site have over time hardened and formed compact deposits. The highest deposit is of the kgosing ward which measures about 1.2 m in diameter and 40 em in height. All hearths are present, even though some may be very shallows deposits. During the 1982 initiation of Mangana, a plan of tlIe camp was drawn complete with branch enclosures that are otherwise bumt down at tlIe end of the ceremony. The map gives all details of the camp layout (Grant, 1984:11) and no further study of tlIe site layout would add much to the already existing information from the map. Motlhwareng is an open space situated about two kilometers Northeast of the main kgotla. It is where all bogwera participants gather on the last day before departing for the bush. It is an assembly area where [mal examination of initiates takes place. Those found to be seriously ill may be sent back home or to the hospital until they recover. Transport arrangements for students and workers are made for them to join the rest of the participants on weekends. Motlhwareng appears to be tlIe only initiation site in and outside Mochudi that has been used continuously from the time when Mochudi was established. It therefore has been used for over one hundred years but with only two initiation features, a motlhware tree and an older log of the same tree. The ritual significance of this tree is still unknown. In the 1980s a new motlhware tree was planted next to the log of the older one that had died earlier on and the two are next to each other. S'tlIobong is an area around Lentswe's residence. There is an open gro~ where ~s often spend some time singing and receiving lectureS during bojale cerem~mes. There IS no material culture nor any features associated with bojale at S 'thobong and It appears to have been used because of its openness. There are no dates of when it was ~ used .for boj~le ceremonies but it has been used in recent times. Seboeng is another Site associated WltlI bojale ceremonies. It was initially on the outskirts of tlIe village on the ~theastern .side but now the village has expanded beyond this place. DikgaJaopa (the last mght of bOJale) 155 used to be held at Seboeng next to Kgosi Isang' s field It was last used as an initiation site bv Mantshakgosi in 1979 and since then dikgalaopa have been held at the show ground Mmaesoke is a place on the outskirts of Mochudi on the western side which was proposed by Mrs. Kgafela to be used for bojale since the I 980s. All bojale sites mentio~ed lack initiation features and are therefore not identifiable archaeologically. Bogwera sites have features that may identifY them archaeologically. Rituals and the use of space There is a clear contrast of the use of space inside or outside either a house, a compound, a domestic space and even a village. For instance the cultural practice of Kgatla funerals was that during funerals women were expected to sit inside compounds while men sat on the outside of the compound of the deceased. Any decisions or announcements that were to be made were done separately inside and outside the compound or yard Men built their own fire outside the compound in their assembly area while the women's assembly area was without fire. The only time women came into contact with fire was when they were preparing tea or food In pre-Christian times there was a clear contrast in terms of places for burying people of different sexes. In the absence of graveyards, children were put inside pots and buried in the house, women were also buried inside houses while men were often buried inside cattle byres or fields. The Bakgatla believed that a woman would continue guarding and working in the house in her afterlife and likewise a man was to continue taking care of his animals or crops. Thus, dead people were buried in areas or spaces that they had occupied while alive. In life women were restricted to the house while men roamed around freely in the bush looking after cattle. During their initiation ceremonies, the Bakgatla girls were initiated around the villa~e. Girls performed their initiation activities at night when people were asleep to avoid bemg seen, but contrary to that boys were literally taken away from the village into the bush and returned only when all activities had been performed. Some informants have argued that boys needed to be away from the village to ensure that they be completely barred from sexual contacts before the circuntcision scars were healed. However, there are those mephato whose initiation procedures did not include circuntcision but who nonetheless spent some time in the bush during their initiation ceremonies and therefore the reason mentioned above seems inadequate. Lincoln (1981) believes that the resolution of contrast or opposites always involves a move from separation to unity and from limitation to totality. Before initiation people :u:e identified as individuals but after going through the ceremony they are identified by therr mephato which groups a large number of individuals into large units. By becoming a part of the process of contrast (in terms of the use of space) an immature boy or girl becomes an adult as the nature of his or her very being is radically transformed (Blacking, 1969; Lincoln, 1981). Lincoln (1981) argues that the contrast between what initiation accom~lish~s for men and for women is graphically presented in spatial terms of their respectlve ntuals. For men initiation is an installation to office (e.g., mephato were used as armed fo~ces and they also provided free community labour) and a change from one level of ~u~on~ and ~sponsibility to a higher one (e.g., only initiated men were allowed to J>3I'tl:Clpa1e ~.deciSlon making on tribal issues at the kgotla) and as such it must involve a spatial transltton or change in residence. For a woman there is no installation into office and therefore her initiation needs no special change. A woman remains where she had always ~~ in terms of authority and office and this is why she is initiated within her nonnal hvmg space rather than doing it elsewhere. The importance of initiating a Mokgatla girl appears to have been noticed by women mostly, while initiating a Mokgatla boy was a 156 thing emphasized by men and women equally. Male initiation was clearly the whole tribe's pride but female initiation was more a women's pride. When men arrived at the kgotla from the ~ush they were met by happy crowds of men and women wearing attires and masks of all kinds to decorate themselves. Even women did wear elaborate attires on such occasions. But, when girls arrived at the kgotla very few men wore their elaborate attires to meet the girls. Male initiation was much more elaborate, complex and resource consuming than female initiation. Change in the authority level of initiated men affected both men and women while authority of initiated women was felt more by other women and less by men. Male initiation pattern is that in which initiates are carried away from the village into the bush (separation), where they are subjected to various ordeals and given instructions (liminal rites), after which they return to the village where they are received as adults (re- incorporation) (Lincoln, 1981). The female pattern is such that initiates never leave the boundaries of the village during their liminal periods. For girls it is not a time of separation but only a time of limited social contacts. Their spatial locus is not changed, only the nature of their activities change. Instead of the three stages of separation, liminal rites and re- incorporation, bojale stages are those of enclosure (within the village), magnification (of the initiates social role) and emergence (girls emerge as women) (Blacking, 1969). Although Lincoln's study was based on women's initiation rituals of the communities of the Tiv, Tiygar, Navajo and Tukuna, his findings and conclusions are clearly applicable to the Bakgatla' s case and it may probably be applicable to other Bantu societies. Another contrast worth looking at is the importance of fire as a central feature in bogwera camps and its absence in bojale places. The first fire built in bogwera camps is that of the kgosing ward (sebesong) and other wards may take an already burning wood from the kgosing hearth to light their own fires. The kgosing fire has to be constantly alight and in fact there may be one or two initiates whose role is to ensure that the kgosing fire never dies. When other people go out hunting they (fire custodians) may remain behind and are instructed by old men never to let the fire die (Roberts & Winter, 1915; Monnig, 1967; Schapera, 1978). The kgosing fire, which was also a central fire for the whole bogwera camp was ritually significant. Initiates going for hunting carried with them a burning wood (Roberts & Winter, 1915). In bojale schools there was absolutely no fire. This was probably because girls stayed at home throughout their initiation seasons but on the last night before graduation (dikgalaopa) they stayed the whole night singing in an open ground Fires were built but only for older women who had undergone the ceremony previously. Fire was not so important nor significant in bojale schools as it was in bogwera schools. This contrast of absence and presence of fire may probably be explained in terms of Tswana ritual beliefs even though none of the informants was able to give an answer for this contrast. The Batswana believe that women, and especially girls in puberty, are ritually 'hot' and 'unclean'. Only older women who have gone past the chil~ ~g age ~e thought to be cool. The reason for not using fire in bojale may be to avOId wttates commg into contact with ritual fire since they are already ritually hot. They need to be cooled down and hence they perform initiation activities at night in winter when temperatureS are at therr lowest. Identification of bogwera sites .. In the examination of Kgatla initiation sites it has been observed that bOJale ~tes are not physically marked by permanent features and therefore ~ey are not likely to be recognizable archaeologically. On the other hand bogwera SItes have been ~own to be marked by more permanent physical features such as hearths. ~~ra SItes can be identified by a concentration of hearths isolated in the bush. There IS a difference between 157 hearths on circumcision camps and of ordinary bush camps. Camps made in the bush were occupied for a short time, often for less than a week and the ash deposits are therefore very thin. Unlike in circumcision camps, hearths in bush camps are not often well preserved because they are not too thick and also because they are left in the open where transformation processes would work on them rapidly. Circumcision site hearths may be encircled by stone slabs as is the case at Tsope and they may also be thick enough to form compact deposits that are more durable, as at Kgogometsong. Bush camp and circumcision hearths are similar in that they are found alongside streams or next to small dams. In recent years boreholes have been used more frequently because they are much more reliable tban natural water bodies. The hearths are also much bigger because there would have been many more people sitting around the same fire than at home. In this respect bogwera hearths are different from domestic hearths. Bogwera fires were not necessarily meant for cooking, they were also meant for congregations and hence they were much larger and more compact. Fires in camps were also used for lighting at night and this meant that they had to be big to produce enough light. Bogwera camp sites are often not associated with permanent occupation materials like house remains. They are often found away from villages and they were not built using durable materials. Tree branches were used to build enclosures and when camps were relocated those branches were burnt down leaving no traces behind Thus, apart from hearths there are no other forms Qf architecture indicative of occupation on bogwera sites. There are also very low artifact densities on bogwera sites. Initiates traveled lightly and did not carry out many tasks at the camp that would produce artifact discard patterns. On these sites there are very low pottery densities. Pottery found is mainly in the form of pot sherds that would have been accidentally broken during the ceremony. There would not be many complete vessels because they were the personal property of individuals who had taken them from home and therefore had to bring them back at the end of the ceremony. There would probably be very few complete vessels that would have been left behind Participants in bogwera schools may engage in wood craft but it is highly unlikely to find wood debris from craft making. Leather working was also common in bogwera schools but like wood it would not be possible to find traces of it in the archaeological record. Leather easily decays and is fed upon by insects making it difficult to preserve. No major metal work is done during initiation ceremonies because metal working itself requires much more time, labour and equipment. Only minor works may be done, for instance making and sharpening knives which would not be traceable in the archaeological record The faunal record of bogwera sites would consist of both domestic and wild fauna but the~e. ~o~d be more wild fauna than domestic ones. Hunting was one of the major actIVitIes m bogwera ceremonies and the animals killed were taken to camps for consumption. Hunting was not only meant for eating but also for obtaining animal skins to be used for making clothes, shoes, blankets and other items. Wild animal bones on these sites therefore would not only be of animals that were eaten but of animals that were e~loited. for their ~ such as genets, lions, leopards, jackals and foxes. There would be a WI~ vanety. of ammals consumed on the sites ranging from small animals like hares, ~ls, gumea fowls to big animals like antelope and cattle. If preservation conditions were I~ there would be a good indication of floral remains on bogwera sites. These would mclude plants exploited for food such as wild berries (moret/wa and mogwane) and ~~ trees used for ~g wooden implements. Wild plants exploited for food are a good mdi~or fo~ seasonality because initiation ceremonies are held during specific seasons, in ~cuIar WInt~. Plant seeds would be mainly preserved as charred remains if they do get mto contact With fire. Tree species obtained for making crafts would also indicate 158 seasonality i~ that they would have to be in a particular stage of thickness and dryness. Trees explOIted for firewood would obviously be preserved in the fonn of charcoal fra~ents and they wou.ld be the most well represented in the archaeological record. It IS a common practice among many Tswana societies to demarcate some areas within and around their villages for ritual and religious activities. Such areas are often laden with myths and beliefs that create fear among people, who thus treat them with respect. Among the Bakgatla such places would include initiation sites, obvious archaeological sites such as Modipe hill and Matsieng, and some parts of the Notwane river. People would only go to these places to perform ritual activities, It was believed that when encountering such places one had to do an act that would please the gods in the sacred area (Schapera, 1960). It was a common practice to carry small stones which were piled up at some points as a token of appreciation and fear for the gods. It is said that at Tsope and some other places when people encountered the sacred areas they took with them stones which they put on previous collections to honour the gods. A lot of piles will be created if the sites were frequently visited and this may result in stone cairns similar to the ones at Kubu and Kaitshe (Huffman, I 996).At the time of investigation the ones at Tsope had been totally destroyed. It is said that there were three of them made mainly by people herding livestock in the area. On the basis of the descriptions of initiation sites and on the basis of what people do in sacred areas it appears unlikely that the cairns at Kubu and Kaitshe are necessarily indicative of initiation. I would argue that they are indicative of places demarcated for religious purposes. The density of the cairns probably imply that they were frequently being used. A comparable group of stone cairns is found in the Sua Pan area. Mike Main has recorded a number of stone cairn fields along the Mosu escarpment (Reid & Segobye, 2000) and the basic characteristic of these features is that they do not have any materials in association. Further study in the Sua Pan area is needed to establish the possible use of the cairns. Within this area there are also Zimbabwe tradition sites at Kaitshe, Toronju, the Kharna site and Tapana that Huffman suggests are settlement sites associated with the cairns, therefore implying that the cairns are initiation features of the Zimbabwe tradition people. But at the moment there is not enough evidence to link the cairns with these settlements because there are no cultural materials on the cairns. Current research in the area (Reid & Segobye, 2000) suggests that the area was occupied between AD 900 and 1500, the most recent being occupation by the Bakalaka, Ban~ato ~~ Basarwa .at Moojane ca. 1800. Stone cairns have been found at Moojane but theIr aSSOCIatIonto the SIte is still unknown. This recent research does not give positive identification of the cairns as initiation features. There is basically not enough evidence to make us conclude that the cairns are initiation features. However, it is not sufficient to disregard Huffman's conclusions on the basis of only one society's initiation evidence. In order to cross check and verify the implications of stone cairn fields of the Sua Pan area we would need to do more survey and excavations. Conclusion This study has been able to describe criteria that archaeologists .m.a~~ able to use to read the archaeological record of ritual sites. The way in which an InItiatIon ceremony am~ng any society is carried out determines the material culture, featIITes and other non-phy~Ical items (e g songs) that are used While in seclusion Bakgatla men may camp at VarIOUS .. , . 'lability of water places in the district and the length of time of campmg ~ds on the. aVaI while some and wild animals Camping sites may be occupied for as lIttle as one mght ::1 ailabl . ft n built using Y av e may be occupied for over two weeks. Camps were 0 e bum do materials, especially tree branches. When moving out of a camp branches were t wn 159 and as a result it was difficult to map any camps outside Mochudi because the scanty occupation deposits were quickly eroded by wind and water. However. circumcision camps were occupied longer, resulting in thick deposits of ash that are the only remnants of male initiation schools. These ash deposits will be localized. Female initiation schools do not have any physical evidence of site usage since initiates stayed at home and only went for initiation sessions at night. Girls were not put in seclusion but were restricted to few social contacts during their initiation ceremonies There is not much communal material culture associated with initiation ceremonies of both boys and girls. Attire worn was personal property and their preservation depended on how much their owners cared. Much traditional attire is made from organic materials (i.e., plant roots and animal skin) and therefore their archeological durability is not guaranteed. An observation made is that there appears to be a relationship between the use of space and gender in rites of passage that the Bakgatla practice. Areas of male domination are usually on the outside while women conquered the inside domestic space. Generally most of the analysis and examination of ideas and issues that came out of this research tend to dwell more on men because male initiation is more likely to leave an archaeological trace while women use hardly any archaeologically durable materials. There is therefore a social selection for men as potential archaeological record producers. The case study of Bakgatla demonstrates that ritual sites need to be analyzed from different points of view, for instance the society's use of space and what that use of space implies in terms of gender relations and also from the point of view of social organization where an initiation site's layout is explained by the way in which a society is organized. Another key factor to look at is a society's ritual beliefs and how that society relates itself to, and interacts with ritual sites and ritual objects. Appendix Regiments initiated at Mochudi REGIMENT LEADER PARENT TEACHER ? Makgal0 REGIMENT Senewe Kgamanyana ? 1874 Matlakana Lentswe Kgamanyane Masoswe ? ? Maisakorna Mapidiwe Kgamanyane ? 1880 Makoba Ramone Kgamanyane Mangana 1884 Majanko Modise Kgamanyane Matukwi ? Matshabatau Mojaki Kgamanyane ? Majekere Nkate Kgamanyane 1892 Mantwane Mochele Kgamanyane Mafatlha 1895 Mabusapel0 Nthakeng Kgamanyane 1902 Makuka Kgafela Lentswe 1911 Machechele Matlakana Matshabatau Isang Lentswe Makoba Mantwane 1915 Mafatshwana Ofentse Lentswe Majanko Makuka 1920 Maatlametla Maramane Pilane 1922 Madirna Radikolo Lentswe Mantwane Machechele 1928 Machama Molefi Lentswe Makuka 1929 Matshego Mafatshwana Motlatsi Pilane 1939 Magata Mmusi Lentswe Machechele 1947 Mafui Madima Ramono Pilane Mafatshwana 1955 Mangope Machama Mokoke Lentswe Madirna 1%2 MaIwelakgosi Magata Tshire Lentswe 1%3 Mathulwa 1973 Lincwe n Kgafela Machama Maisakorna Marui Khukhwanana Pilane Machama 160 1975 Masoswe Ramonnye Kgafela Magata Mangope 1976 Madingwana Phulane Pilane Mafrri Mathulwa 1978 Mabusapel0 Seipati Pilane 1979 Mantshakgosi Mmalebotse Lentswe Malomakgomo KubePone Mangope Masoswe 1980 1981 Majekere Seingwaeng Kgafela Mangana Kgafela Kgafela Mathulwa Madingwana 1982 Bakgatla Kgafela Masoswe Malomakgomo 1985 Manoga Mmusi Kgafela Madingwana Mangana 1988 Matukwi Notes Ms. Morongwa Mosothwane is currently a postgraduate student at the University of Pretoria. Her research essay was completed in 1999 under the supervision of Alinah Segobye. The original includes a number of photographs which have been omitted for space. The text was lightly edited References Oral Sources Female informants WARD SURNAME AGE REGIMENT NAME Manamakgotla Gouwe 89 Lechama Marea Kgosing Kgafela 54 Leisakoma Mmarakau Kgosing Pheto 67 Lengope Mogaritshi Mabodisa Mfetane 97 Lechama Mmatsatsi Manamakgotla Sello 96 Lechama Kgatheng Kgosing Pheto 71 Lengope Matlhodi Kgosing Pilane 73 Lefiri Motsatsi Morema Bokole 88 Letiri Mmathari Manamakgotla 35 Lelomakgomo Sampo Gouwe Morema Lepiri 74 Lefrri Marothodi Mabodisa 90 Lechama Motlatsi Diswai Mabodisa Podile 99 Ledima Shanto Lethulwa Manamakgotla Mosothwane 59 Nkidi Lebusapelo Morema Molefe 61 Sakatsane Ledima Morema Mmamokgethi Kitsi 100 Legata Morema Lesane 78 Sebitse Letshego Tshukudu Maiko 81 Lemme Lekhama Tshukudu Molefe 80 Mmantho Lefrri Morema Lebekwe 70 Mmadikeme Legata Manamakgotla Moje 68 Manamakgotla Mokhuthe Lethulwa Sebakwe 60 Manamakgotla Mmasengwe Lengope Mokgatle 75 Manamakgotla Lesedi Lengope Ntshole 72 Manamakgotla Matlakala Lechama Modisana 95 Manamakgotla Bogadi Legata Sebako 73 Manamakgotla Motsei Lefrri Tladi 78 Mmamonala Letshego Manamakgotla Tshetlhane 72 Manamakgotla Rabeka Legata Pheto 78 Kgosing Motlatsi Legata Selemogwe 78 Seipora Lethulwa Tshukudu Mmutle 63 Fenenke Lebusapelo Kgosing Pilane 40 Maruapula 77 Lengope Kgosing Tietseng Kgarnanyane Kgosing 88 Legata Sabona Mmale 161 Male informants NAME SURNAME AGE REGIMENT WARD Kgosi Lincwe II Kgafela 60 Lethulwa Kgosing Sandy Grant Lethulwa Modise lladi 80 Lefrri Ntlhe Moje 65 Legata Manamakgotla Ramono Mongale 82 Lefrri Manamakgotla Pone Kube 61 Ledingwana Kgosing Mpuwe Makakaba 84 Legata Kgosing Tsae Mosothwane 84 Legata Mabodisa Moreri Tshite 82 Legata Manarnakgotla Ramfi Seoke 64 Lesoswe Mabodisa Lentswe Kgarnanyane 85 Lefrri Morerna Rarnontie Segwati 95 Lediwa Kgosing Thipe Segwati 41 Lengana Kgosing Lobi Kgotele 98 Lediwa Kgosing David Mokgadi 45 Lengana Tshukudu Modise Ntshole 61 Lethulwa Manarnakgotla Thari Bokole Show Lethulwa Morerna Mfolwe 65 Lengope Ditshwene Manarnakgotla Masike 77 Lefrri Bakgatla Morerna Kgafela 22 Lenoga Mmusi Kgosing Kgafela 20 Letukwi Kgafela Kgosing Kgafela 28 Lengana Kgosing Botswana National Archives BNA S217/6, Bogwera and Bojale. Application of proclamation to Bechuanaland Protectorate 1931 BNA S218/1O, Notes on interviews with ChiefBathoeng II 1931/05/21 BNA SI00/2, Medical Officer 1930/05/31 Published Sources Blacking, 1. (1969) 'Songs, dances, mimes and symbolism of Venda girls' initiation schools' (parts 1- 4) African Studies, 28: 3-36; 69-118; 149-200; 215-266 Breutz, P.L. (1953) The tribes ofRustenburgand Pilansberg Districts Pretoria: Government Printer Evans-Pritchard, E.E. (1940) The Nuer Oxford: Clarendon Press Grant, S. (1984) 'The revival of bogwera in Kgatleng _ Tswana culture or rampant tribalism? A description of the 1982 bogwera' Botswana Notes and Records, 16: 7-17 Grant S. & L. Kgafela, (1989) 'The relevance of the past in the present in Mochudi, Botswana _ initiation and museum ' World Archaeological Bulletin. 3 :5-11 Huffman, T. N. (1996) Snakes and Crocodiles Johannesburg: Witwatersrand University Press Kollars, P. (1991) 'Bojale 1990',73-76 in Tshomarelo ngwao _ The Museums of Botswana celebrate twenty-five years of independence Gaborone: Government Printer Lincoln, B. (1981) Emerging from the Chrysalis: studies in rituals of women 's initiation Cambridge, Mass.: Harvard University Press Monnig, H. O. (1967) The Pedi Pretoria: Van Schaik M~tlotle, N. P. (1998) 'Bojale - girls initiation school' The Zebra's Voice, 25 (2): 19-21 ., Reid, A & AK. Segobye (2000) 'Politics, society and trade on the eastern margins of the Kalahan South African Archaeological Society Goodwin Series. 8:58-68 Richards, A I. (1956) Chisungu London: Faber Roberts, N. & CAT. Winter (1915) The kgoma or initiation rites of the Bapedi of Selrukuniland' South AfricanJournai of Science, 12: 561-578. 162 Schapera, 1. (1936) The contributions of Western civilization to modem Kxatla culture, BNA BNB 388 Schapera,1. (1942) A Shorl History of the Bakgatla Bagakgafela of Bechuanaland Protectorate Cape Town: University of Cape Town Schapera,1. (1955)A Handbook of Tswana Law and Custom London: Cass Schapera, 1. (1956) Government and Politics in Tribal Societies London: Watts Schapera, 1. (1960) 'Christianity and the Tswana', 489-503 in S. Ottenberg & P. Ottenberg (OOs.) Cultures and Societies of Africa New York: Random House Schapera, 1. (1978) Bogwera: Kgatla Initiations Mochudi: Phuthadikobo Museum van der Vliet, V. (1974) 'Growing up in traditional society', 211-234 in W.D. Hammond - Tooke (ed.) The Bantu Speaking Peoples of Southem Africa London: Routledge and Kegan Paul Willoughby, W. C. (1909) Notes on the initiation ceremonies of the Becwana, BNA BNB 370 163 .I '. I r I r-. • I r .~et .. ~x -' • ...\O~ -' ~fle ~.J _\- _-' • \ ,-, , - .... ,-- -- \ .......... .... ( .... _ .... I -' ,,7 I , ----~-,,.._I / \ Mochudi -' I -' " I , I I ,r Borehole • \ ) ( / 40km ( Fig. 1. Mochudi and nearby boreholes used as initiation sites. 164 200m / •• • /' N / / ,/ /' / / •• • •• /' • • • • • Hearths Fig. 2. The initiation site at Tsope with an enlarged view of a hearth. 165