Farming people did not inhabit the Greater Johannesburg region until the Late Iron Age. Then, beginning in the 15th century, BaFokeng dominated the landscape. A few other Sotho-Tswana people, most notably BaKwena, also lived in the region. Large stonewalled settlements of both BaKwena and BaFokeng characterised the troubled times of the difaqane/mfecane at the end of the 18th century. Mzilikazi, however, depopulated the region in 1823; and so, the land appeared empty when Voortrekkers arrived.
To understand this brief outline, it is necessary to start at the beginning of the Iron Age, 1800 years ago.
The paper was written for SAHO in 2010 by Prof TN Huffman from the Archaeology Department at the University of the Witwatersrand.
Mixed Farming and Climate
About 1800 years ago, Bantu-speaking people brought a new way of life to South Africa. For the first time, people lived in settled villages, cultivating such crops as sorghum, pearl millet, finger millet, ground beans and cowpeas. These new farmers also herded cattle, sheep and goats, and so they are called agropastoralists. Because they also produced their own metal tools, archaeologists call this time the Iron Age.
Early Iron Age (AD 300-900) people chose to live in broken country that had sufficient woodland, water and soils that could be cultivated with iron hoes. The climate also had to be sufficiently warm and wet to meet the biological requirements of the domestic crops. Sorghum and millets, for example, need a minimum of 350 mm of rain in the summer growing season. Essentially, these crops cannot grow with less than about 3 mm of water a day: 50 days for the two millets and 75 days for sorghum. To meet these requirements, the minimum annual rainfall needs to be about 500 mm. Furthermore, night-time temperatures should not drop below 15ËšC.
The Magaliesberg valley north of Johannesburg was ideal, and farmers settled there by the 6th century. The best-known site is at Broederstroom, west of Pretoria. The Johannesburg area, however, was generally too open, dry and cold for subsistence farming. During the Middle Iron Age (AD 900-1300), the climate was better than it is today, that is, it was warmer and wetter. But better land was available elsewhere, and agropastoralists did not settle in the Johannesburg region until the Late Iron Age (AD 1300-1840).
Language and the Late Iron Age
According to linguistic research, Sotho-Tswana and Nguni speakers lived in East Africa during the Early Iron Age. Both languages are divisions of Eastern Bantu that in turn is a division of the larger Bantu language family (All Bantu languages originated in West Africa where Nigeria and Cameroon come together).
Archaeologists trace the movements of these and other people through characteristic material-culture remains, usually ceramic style. It is possible to use ceramic style this way for two reasons. First, language is the principal vehicle for thinking about the world and transmitting those thoughts to others; and so, there is a vital relationship between language, worldview and material culture. Secondly, provided that the makers and users belong to the same material-culture group, the distribution of a ceramic style represents the distribution of a group of people who speak the same language.
According to the ceramic evidence, Sotho-Tswana and Nguni speakers moved south into southern Africa between about AD 1100 and 1300. Generally, Nguni occupied the eastern regions, while Sotho-Tswana moved onto the plateau, starting in the Limpopo Province.
After a while, the first Sotho-Tswana split into two clusters: a Western cluster (that today includes BaHurutshe, BaKwena, BaKgatla, BaNgwaketse and BaNgwato) centred in the present-day Northwest Province; and a Southwestern cluster (including BaRolong and BaThlaping) that inhabited the region from the Magaliesberg to Potchefstroom, including Johannesburg. Radiocarbon dates place the pottery (called Olifantspoort after the site where it was first recorded) between about AD 1450 and 1700.
A second branch of Sotho-Tswana moved through Natal and settled in the Free State near the hill called Ntsuanatsatsi. This branch forms a third, Southeastern cluster. According to well-established oral history, these people were BaFokeng. Although the climate was warmer and wetter than today, the highveld was nevertheless relatively tree less. Consequently, BaFokeng used stonewalling to mark boundaries where wood and reeds would have been the norm elsewhere.
This early walling is called Type N (after Ntsuanatsatsi). It is characterised by a stone circle about 20 to 30 meters across inside a larger circle some 60 to 80 meters in diameter.
This and other Sotho-Tswana walling represents what archaeologists call the Central Cattle Pattern. In this pattern the centre, the domain of men, contains the men’s court, metal working area and cattle kraal. Men are usually buried in the kraal, and it is a place to propitiate male ancestors. An outer ring of houses forms the residential zone and the domain of women. This zone is divided into the individual households of married women; each includes a front courtyard, kitchen, sleeping hut and backcourt with grain bins and other storage items. Women are usually buried in the back courtyards of their households, while children, depending on their age and other circumstances, were usually buried in the front courtyard.
In the case of Type N walling, the larger outer circle marked the homestead boundary, while the inner circle marked the central cattle kraal. The occasional straight low wall between the two divided the residential zone into separate households. If present, small stock enclosures were incorporated in the outer perimeter wall.
Settlement patterns everywhere are the result of a society’s worldview. In this case, the Central Cattle Pattern is associated with Eastern Bantu speakers who were patrilineal (that is they traced their blood from their father), had male hereditary leadership, believed that male ancestor spirits played a positive role in daily life, and they exchange of cattle for wives. Bridewealth in cattle, in fact, was a major institution that established marriage ties, status and political power.
This short discussion of settlement organisation, language and climate provides a background to the events and way of life of people living in the present-day Johannesburg area during the last 500 years.
The early history of Johannesburg
In the 15th century BaFokeng people, using the early type of walling, spread north across the Vaal. Type N sites are on record near Balfour, in the Suikerbosrand, Vredefort Dome, Pretoria and Greater Johannesburg area. For Johannesburg, some of the best examples occur in the Klipriviersberg to the south. The associated pottery is called Uitkomst (after the name of a cave where it was first found). Radiocarbon dates place this first walling with Uitkomst pottery between about AD 1440 and 1665.
This period was warmer and wetter than today, and early BaFokeng were able to cultivate sorghum and millets in river valleys, such as the Klip and its tributaries. Most settlements were built on hill slopes where stone was readily available and in close proximity to the fields. This last point was true for the vast majority of Iron Age agropastoralists: even though cattle were immensely important, settlements were usually located in terms of agricultural priorities.
The distribution of Type N settlements follows a dispersed homestead pattern, that is, most settlements consisted of individual homesteads spaced a few hundred meters apart. Each contained some 30 to 50 people, half of whom would have been children. These were the homesteads of ordinary men.
Larger settlements marked the capitals of chiefs and district leaders. ‘Big men’ lived in big settlements because political power was based on the unequal distribution of wealth. Throughout the Late Iron Age in southern Africa, chiefs were the wealthiest people in their nations. They usually had more cattle than anyone else, accumulating them through death dues, court fines, forfeits, tribute, raids and the high brideprice of their daughters. Chiefs used their large herds to establish allegiances through loans and political alliances through bridewealth in cattle.
Besides formal marriage alliances, a chief might receive wives as tribute, and he usually had more fields around his capital than anyone else. In addition, there were also a number of other fields cultivated for him as a form of tribute. In fact, a nation normally wanted its chief to be wealthy so that he could function properly. As ‘father’ of the nation, the chief was expected to feed his people in time of famine, to support widows and others who could not support themselves, to loan cattle to poor men and to supply refreshments to visitors in the capital.
To help them function, chiefs maintained a number of court officials, such as guards, messengers, retainers and councillors. The capital was active not only because it was the national centre but because more people lived in the chief’s district than elsewhere. Thus, a capital needed to be large to accommodate all the activities as well as the relatively high number of families, officials and followers who lived there.
Chief’s settlements were made up of a number of individual homesteads aggregated together. A normal chief would have something like 300 to 500 people living with him. He would have the largest homestead and therefore the largest Type N arrangement. The other homesteads were placed in a special order around the chief in the same way that the houses of different wives followed a special order inside each homestead.
The settlement of a chief has not yet been recognised in the Greater Johannesburg area, but there may have been one in the Klipriviersberg, where many Type N sites occur. If the Klipriviersberg area was the core of one chiefdom, then typically, it would have been separated from the next chiefdom by empty land.
The degree and nature of interaction between BaFokeng and Southwestern Sotho-Tswana is unknown. By the mid 17th century, however, some BaFokeng had moved north to the Springbok Flats and Waterberg where they appear to have merged with Western Sotho-Tswana: typical pottery of both groups has been found together in a single homestead. In the Johannesburg area, though, BaFokeng appear to have been dominant. BaFokeng living today near Rustenberg may well be direct descendants of the original movement from Ntsuanatsatsi.
All agropastoralists appear to have left Greater Johannesburg between AD 1670 and 1780 because the climate became cooler and drier. When conditions improved 100 years later, Sotho-Tswana farmers once again lived in the area.
Some of these people were BaFokeng, as their characteristic Uitkomst pottery shows. But now the type of walling is more complex: the centre contains more cattle kraals, the residential zone has more divisions and the outer perimeter includes arcs to demarcate back courtyards. This more complex pattern is called Klipriviersberg and many examples cover the hills in the Klipriviersberg.
Another type, known as Molokwane, is found in the Magaliesberg and Suikerbosrand. In each area, oral history and archaeological data identify this type with Western Sotho-Tswana, and BaKwena in particular. This walling differs from the other two in that small stock enclosures stand in the central area, rather than the residential zone, and arcs in the outer wall mark the back courtyard of every household. Presumably, BaFokeng introduced the use of stonewalling to BaKwena.
Both Molokwane and Klipriviersberg settlements were built with high walls on hilltops in defensive positions. Furthermore, both had an aggregated rather than dispersed pattern: individual homesteads were rare, and most settlements were clustered into small towns. Similarly, capitals became enormous towns, such as Molokwane, the BaKwena capital west of Rustenberg, and these towns housed many thousands of people.
The location and large sizes were Sotho-Tswana responses to the military tension of the troubled time known as the difaqane, or mfecane. Among other factors, populations multiplied as rainfall increased in the second half of the 18th century and people added maize to their farming strategies. Maize could be successfully grown because of the increased rainfall. Furthermore, maize is easier to cultivate, and its yields are higher than sorghum and millets. Characteristically shaped grindstones show that BaFokeng and BaKwena people cultivated maize in the Greater Johannesburg area by this time.
Oral history and tree-ring studies show that the high rainfall was followed by a severe drought at the beginning of the 19th century. This drought caused widespread famine and increased competition for resources that led to an abnormal period of strife: Nguni attacked Nguni, Nguni attacked Sotho-Tswana and Sotho-Tswana attacked Sotho-Tswana.
As a result, some groups lost their identity, and individual survivors were forced to form new alliances. For example, one BaRolong homestead is located in the Klipriviersberg inside a BaFokeng community, while BaFokeng pottery has been found in BaKwena settlements in the Suikerbosrand.
Sotho-Tswana control, and probably occupation, came to an end in Greater Johannesburg in 1823 when Mzilikazi conquered the area. As is well known, Mzilikazi was a Khumalo leader who left Natal to avoid Chaka’s wrath. Mzilikazi first established his headquarters near Heidelberg before moving to Pretoria. It was his policy to empty the border-land of people and to keep his followers close to the capital - including Sotho-Tswana who paid tribute. Many Group II settlements were burnt to the ground at about this time.
Mzilikazi moved west from Pretoria to Mosega near Zeerust, but he still claimed his old land. Thus, he sent warriors within two days to attack the Voortrekkers after they had crossed the Vaal in 1838. That same year, the Voortrekkers, with the help of BaRolong, forced Mzilikazi to leave South Africa. Within a short time, the Boers declared a Republic and started a new chapter in the history of Greater Johannesburg.