Mapungubwe (South Africa)

African Iron Age Polity in South Africa

View of Mapungubwe Hill from the Terrace
View of Mapungubwe Hill from the Terrace. bistandsaktuelt

Mapungubwe is an African Middle to Later Iron Age site (AD 900-1300), located in the Mapungubwe National Park, in the Shashe-Limpopo river basin, Limpopo province of South Africa, adjacent to both Zimbabwe and Botswana. A stratified population is in evidence at the site, with elites residing at the top of the sandstone outcrop on Mapungubwe Hill, and non-elites living on the southern terrace at its base.

Mapungubwe was the center of a polity, referred to in the literature as the Mapungubwe Landscape, with a religious leader based at Mapungubwe Hill and a sphere of influence extending to the Kalahari desert and the east coast. Settlements that were part of the Mapungubwe Landscape include Mtanye, Mutshilachokwe, Princess Hill, Skutwater, Weipe, Little Muck and Mmamgwa Hill. The economy was based on herding domesticated cattle; pearl millet, peas and sorghum agriculture; and participation in the vast Indian Ocean trade network. About 5,000 people lived at Mapungubwe at its height ca 1250 AD.

Chronology at Mapungubwe

  • AD 350-450, first occupation
  • AD 450-900, Early Iron Age (mostly a hiatus at the site)
  • AD 900-1000 Zhizo Phase or M1 Phase
  • AD 1000-1220 Leopard's Kopje/K2 Phase or M2, seat of regional power at K2 Village
  • AD 1220-1290 Eiland/Mapungubwe Phase or M3, power at Mapungubwe
  • AD 1290-15th century AD power shifts to Great Zimbabwe, some evidence of occupation continues

    Gold Graves

    Approximately 27 graves have been identified on the top of the hill. Three of them were looted by the original discoverers, although the majority of the objects were eventually recovered. The "chief's grave" or "original gold grave" was reportedly covered by large square possibly polished stones.

    Based on minimal bone fragments recovered from the looted site, this individual was a young adult, perhaps 25-45 years of age. Artifacts recovered from the shallow grave included gold funerary objects including a sculpture of a gold rhinoceros and a gold bowl, and thousands of glass and gold beads and bangles, and a few ceramic pots.

    The "sceptre skeleton" was a young adult man, buried in a sitting position with a gold sceptre and a second, fragmented gold rhinoceros sculpture. The "gold skeleton" is of an older individual, buried with more than 100 bangles of coiled gold wire and about 12,000 gold beads.

    Living at Mapungubwe

    Mapungubwe contains some of the earliest known evidence for gold, bronze and brass casting in Africa. Among iron wares produced at the site are delicate iron wires, fabricated using the 'strip-twisting technique', also used in decorating pendants. Small copper funnels and conical tubes were made at the site, and scholars believe they were part of the wire manufacturing process.

    Trade goods made at Mapungubwe included glass beads and cloth; a significant Mapungubwe-based trade in salt, hides, ivory and ostrich eggshell beads is in evidence along the eastern coast.

    Chinese celadon ceramics dated to the Song (1127-1279 AD), Yuan (1279-1368 AD) or early Ming (1368-1644 AD) dynasties of China were recovered from the site.

    Over 100,000 glass beads were eventually identified at Mapungubwe, a quarter of which were recovered from one burial. Mapungubwe's bead assemblage included "Dutch Dogons" made in Germany, hexagonal beads from Czechoslovakia, Indo-Pacific beads from India and Sri Lanka, Islamic beads from al-Fustat, and Venetian glass beads: all of these attest to the breadth of the trade system connecting Mapungubwe to the rest of the world.

    Archaeology at Mapungubwe

    Mapungubwe was "rediscovered" in 1932 by a farmer/student who heard rumors about the site and forced a local man to take him to the hill, where an abundance of gold, pottery and beads was identified in the midst of stone architecture.

    First excavated by Leo Fouché of the University of Pretoria in 1933, and again several times over the succeeding decades, Mapungubwe has been systematically studied by a raft of archaeologists including Robinson, Summers, Whitty, Garlake and Huffman. UNESCO named Mapungubwe a World Heritage Site in 2003.

    Archaeologist Rachel King (2011) recently discussed the role of the African game of mufuvha at Mapungubwe, and how it is used to enrich the social and political heritage of Mapungubwe. A board where the game could be played was carved into the sandstone within the main complex of Mapungubwe Hill. Mufuvha, or a version of it, is known throughout the subcontinent of Africa and parts of east and west Africa as well. Historically, the game of mufuvha is played only by men, and traditionally it is associated with gambling and herding or stealing cattle. Although the rules vary from place to place, it always involves capturing pieces to win.

    Reanalysis of beads and celadon pottery found at Mapungubwe and the related site of K2 (Prinsloo et al.) suggests that some of them date to the early Ming Dynasty, suggesting that Mapungubwe cannot have been abandoned until the 14th or early 15th centuries AD, opening up the possibility that these reflect contact via the travels of the Chinese explorer Zheng He.

    Support for the latter dating by Prinsloo and colleagues is found in a recent article (Zink et al.) describing OSL dating of the pottery recovered from Mapungubwe, which also date "too late" for traditional reports of the abandonment from 1290 to well into the 14th or early 15th centuries.


    This glossary entry is a part of the guide to African Iron Age, and the Dictionary of Archaeology.

    Huffman TN. 2008. Climate change during the Iron Age in the Shashe-Limpopo Basin, southern Africa. Journal of Archaeological Science 35(7):2032-2047.

    Huffman TN. 2009. Mapungubwe and Great Zimbabwe: The origin and spread of social complexity in southern Africa. Journal of Anthropological Archaeology 28(1):37-54.

    King R. 2011. Archaeological naissance at Mapungubwe. Journal of Social Archaeology 11(3):311-333.

    Koleini F, Schoeman MHA, Pikirayi I, and Chirikure S. 2012. Evidence for indigenous strip-drawing in production of wire at Mapungubwe Hill (1220–1290 AD): towards an interdisciplinary approach. Journal of Archaeological Science 39(3):757-762.

    Mosothwane MN, and Steyn M. 2004. Palaeodemography of early Iron Age Toutswe communities in Botswana. The South African Archaeological Bulletin 59(180):45-51.

    Neukirch LP, Tarduno JA, Huffman TN, Watkeys MK, Scribner CA, and Cottrell RD. 2012. An archeomagnetic analysis of burnt grain bin floors from ca. 1200 to 1250 AD Iron-Age South Africa. Physics of the Earth and Planetary Interiors 190-191(0):71-79.

    Prinsloo LC, Tournié A, and Colomban P. 2011. A Raman spectroscopic study of glass trade beads excavated at Mapungubwe hill and K2, two archaeological sites in southern Africa, raises questions about the last occupation date of the hill. Journal of Archaeological Science 38(12):3264-3277.

    Robertshaw P, Wood M, Melchiorre E, Popelka-Filcoff RS, and Glascock MD.

    2010. Southern African glass beads: chemistry, glass sources and patterns of trade. Journal of Archaeological Science 37(8):1898-1912.

    Steyn M. 2007. The Mapungubwe Gold Graves Revisited. The South African Archaeological Bulletin 62(186):140-146.

    Zink AJC, Susino GJ, Porto E, and Huffman TN. in press. Direct OSL dating of Iron Age pottery from South Africa – Preliminary dosimetry investigations. Quaternary Geochronology in press.