Field trip to uMgungundlovu

17 Dec 2019 - 09:00
Revisiting the uMungundlovu excavations. Photo courtesy of Steven Kotze.


On 14 and 15 November 2019, eleven academics and museum personnel from five different institutions came together on a field trip to examine the site of uMgungundlovu, the chief ikhanda of the Zulu king Dingane kaSenzangakhona in the 1830s. Among the members of the party were APC chair Carolyn Hamilton and APC research associates Thokozani Mhlambi and John Wright. Here Wright reports.1

Lunguza Mthembu, James Stuart, and other knowledge-makers

Sunday afternoon, 21 March 1909, at the home of colonial official James Stuart in Pietermaritzburg, the capital of the British colony of Natal. Stuart has been in conversation with a visitor, Lunguza kaMpukane Mthembu.

Today I constructed with amabazelo [chips of wood] (for huts) and pieces of bricks & pebbles (for cattle) the Mgungundhlovu kraal under Lunguza’s guidance. I see this differs considerably from Holden’s illustration – see ‘History of Natal’. The illustration on opposite page comprises results arrived at. What Lunguza is most shaky about is the inside of isigodhlo esimhlope & esimnyama. He says umdhlunkulu occupied both and does not know why one was given one name & the other another. He says the ‘white’ consisted of rows of huts in ordinary way & he believes the ‘black’ each had fences dividing certain huts from others. He had occasion to enter the ‘white’ 6 times in all & each time he found Dingana in hut marked Z in the drawing.2

Thus wrote Stuart after ten days of conversation with Mthembu on the latter’s life at Mgungundlovu more than seventy years before. Mthembu had been born in the early 1820s, and had lived in the western borderlands of Dingane’s kingdom in the region governed on behalf of the Zulu king by Chief Jobe Sithole. On a number of occasions the young Lunguza had accompanied his father as a mat-bearer on visits to Mgungundlovu. He had also lived there for a while in the late 1830s after having been called up into the uKhokhothi ibutho. Mthembu was one of four old men who, in the period, 1903 to 1909, discussed with Stuart, in detail and from different perspectives, their experiences while living at uMgungundlovu. The others were Thununu kaNonjiya Gcabashe, Ngidi kaMcikaziswa of the eLangeni clan (praise-name: Magambukazi), and Sivivi kaMaqungo Malunga.

Stuart’s notes of his conversations with these interlocutors fill some 480 notebook pages. In one of the isiZulu school readers that he published in the 1920s, he went on to write up a thirty-page description of uMgungundlovu and of life in Dingane’s court based largely on the accounts they had given him.3 He remains the only writer to have recorded at such length statements on these topics emanating from African eye-witnesses. What was it about uMgungundlovu and the reign of Dingane that so engaged his attention? For me as a historian who has for many years worked with the collection of papers left by Stuart, the prospect of revisiting the site and of being able to discuss this issue with like-minded colleagues was a major reason for joining the field trip.

Stuart was of course far from being the first writer to take an interest in uMgungundlovu. From soon after the ikhanda’s construction, probably in 1829, colonial traders, officials, and missionaries had visited Dingane there, and had produced sometimes lengthy descriptions of their meetings with him and of the activities they observed. Fynn, Isaacs, Andrew Smith, Gardiner, Champion, Owen – their writings form an important element in the archive of documents on uMgungundlovu.

After the defeat of Zulu forces by a Voortrekker commando at the Ncome river in December 1838, uMgungundlovu was burnt down, part of it on Dingane’s orders and the rest, it seems, by the victorious trekkers. The changing – and contested – place subsequently held by Dingane in the historical imaginations of isiZulu-speaking people has only recently become a subject of focussed academic research.4 For Afrikaner nationalist leaders and culture brokers, in the era of colonialism and apartheid, the locality just outside uMgungundlovu where trekker leader Pieter Retief and his party were put to death in February 1838 was long an iconic memorial site. To a lesser extent it remains so today.

After the British had defeated and partitioned the Zulu kingdom in 1879, the region was wracked by civil war. In 1884 the Afrikaner-dominated New Republic, with its capital at Vryheid, annexed a great slice of the former kingdom in return for assisting the inkosi Dinuzulu kaCetshwayo to overcome his rivals. Much of the old Zulu heartland, including the site of uMgungundlovu, was incorporated into a belt of farms marked out for white settlement. The area remained as farmland until very recently. In the late 1990s and early 2000s, under circumstances which I still have to establish, it was bought up from the owners and placed under the jurisdiction of the newly formed province of KwaZulu-Natal. Today it forms part of the eMakhosini-oPhathe Heritage Park.

Archaeologists, curators and digital scientists

In the 1970s the site of uMgungundlovu began attracting attention from professional archaeologists. For them, one of the great drawcards of the site is that when the ikhanda was burnt down the floors of many huts were baked hard and have survived into the present. Once cleared of vegetation, they serve as indicators from which the ikhanda’s former layout can be worked out.  Since the 1970s, teams from various institutions with various purposes have on and off surveyed and excavated there. But there has been no overall management of archaeological research at the site: the result is that the material objects recovered, together with notes and reports written by the excavators, have become dispersed in a variety of locations, and in some cases lost to sight.

Currently the APC’s Five Hundred Year Archive project has a sub-project which digitally reconvenes and curates the material and documentary uMgungundlovu archive. In August 2019 APC chair Carolyn Hamilton visited the KwaZulu-Natal Museum, where some of the archival material is held, to work with Gavin Whitelaw, an archaeologist and chief curator in the museum’s Department of Human Sciences, on sorting out the history of excavations at uMgungundlovu. (See the report, ‘Ongoing FHYA collaboration with the KwaZulu-Natal Museum’, in the APC Gazette of October 2019.)

In October a follow-up workshop held at the University of the Witwatersrand brought together a working group of curators and digital scientists from a consortium of three projects: the FHYA, the Centre for Curating the Archive’s curation of the Bleek and Lloyd collection, based at UCT, and Metse Megologolo, based at Wits. The consortium operates in the Digital Humanities Programme co-ordinated by the Wits Institute of Social and Economic Research. (See the report, ‘First steps towards a supra-institutional archive initiative’, APC Gazette, October 2019.) One of the outcomes of the workshop was a decision for members of the group to visit the uMgungundlovu site.

For their part, the members of the Metse Megologolo project are exploring the possibilities of creating a multimodal archive of African urbanisation, focussed in the first instance on historical Tswana towns. They are concerned with introducing spatial modes of exploration into the digital archive. Having in mind the work already done by the FHYA on the digital archiving of uMgungundlovu, members of the project have decided to explore how the spatial information contained in the documentary archive on this site (in the form of maps, excavation grids and so on) might be used to help develop a spatialized digital platform. The object of the visit to uMgungundlovu was thus to give researchers in the project a clearer idea of the layout of the former ikhanda, the places of provenance of the material archive, the links to documentary descriptions of the site, and the site’s place in the wider landscape.
The trip was organized by Whitelaw and Justine Wintjes, curator of material collections in the KZN Museum’s Department of Human Sciences. In addition to them and the three APC-associated individuals named above, other members of the party were Anton Coetzee, sessional lecturer in Digital Arts and the History of Art at Wits, and a PhD candidate in history of art; Michelle House, who has just completed a PhD in archaeology at UCT; Steve Kotze, researcher in the Local History Museums in Durban; Stefania Merlo, senior lecturer in archaeology at Wits; Frans Roodt, lecturer in archaeology at the University of Limpopo and former curator of the uMgungundlovu site museum; and Hussein Suleman, head of the Department of Computer Science at UCT. In all, the party comprised four archaeologists, three historians, two art historians, a music scholar/composer, and a computer scientist.


‘This great historic capital’

Our first stop was for a brief viewing of the Shaka Memorial and the adjoining visitors’ centre in KwaDukuza situated on the site of Shaka’s final residence. Later on, Whitelaw took us along the road between eShowe and eMpangeni to view – from a distance – the site of King Shaka’s former kwaBulawayo ikhanda. It was built sometime in the early 1820s and occupied until at least late 1827, when the king made the recently erected ikhanda of kwaDukuza, near what is now the town of that name, his administrative centre. Today the site of kwaBulawayo is covered in bush, and little archaeological work has been done there.5 It provides a lofty view of the surrounding terrain, but one has to work hard to imagine the place as the capital of a powerful kingdom nearly 200 years ago.

The site of uMgungundlovu, by contrast, is strikingly visible across the valley of the Mkhumbane stream as one approaches by road. We were fortunate to be there in spring after good rains, with the previous summer season’s long grass cover having been burnt off and the new grass outlining the site’s broad dimensions in verdant green. We were also fortunate to have as our guide Frans Roodt, who surveyed and excavated at the site in the 1980s and 1990s, and knows it better than any other archaeologist.

The ikhanda was situated on a pronounced slope for drainage purposes: generally accepted measurements give it as having been about 580 metres from upper to lower end, and 550 metres across. In two arcs on either side of the central enclosure were between 1100 and 1600 huts, which at any one time housed from 5000 to 7000 people. Mthembu’s statements to Stuart indicate that most of this population would have consisted of young men of the amabutho stationed there, together with temporary visitors. Apart from the women of the king’s isigodlo, and a handful of senior officials, there was no settled population. It was very much an administrative rather than a residential centre.

Several members of the party had read up in detail into the available documentary source material. Both on the site and afterwards, lively discussions took place that helped us give life to what we had seen, and to formulate new ideas and new questions from the perspectives of our different disciplines. The results will become clearer as the project develops. For this historian, most important was to be able to put into clearer context on the landscape the differently angled accounts of their experiences at uMgungundlovu given by Gcabashe, Magambukazi, Malunga and Mthembu.

On 22 March 1909 Stuart finished off his notes on his discussions with Mthembu with the comment:

Not being familiar myself with what has been written by Gardiner, Owen and others, including Isaacs, about Dingana’s day, I could not ask all the questions I might have done,but no-one studying what I have set down independently can fail to get a very clear idea as to what went on at this great historic capital.6

With thanks to Gavin Whitelaw and Justine Wintjes for information and comment.
2 Killie Campbell Africana Library, James Stuart Papers, File 59, nbk. 32, p. 10. See also the published rendering in C. de B. Webb and J.B. Wright, eds., The James Stuart Archive of Recorded Oral Evidence Relating to the History of the Zulu and Neighbouring Peoples, vol. Pietermaritzburg: University of Natal Press, 1976, p. 339.
3 James Stuart, ‘uMgungundhlovu’, parts I, II and III, in uKulumetule, London: Longmans, Green, 1925, pp. 14-45.
4 Nsizwa Dlamini, ‘Patterns and shifts in cultural heritage in KwaZulu-Natal: selected case studies, 1977-1999’, MA thesis, University of Natal (Pietermaritzburg), 2001, ch. 5; Sifiso Ndlovu, African Perspectives of King Dingane kaSenzangakhona: the Second Monarch of the Zulu Kingdom, Cham, Switzerland: Palgrave Macmillan, 2017.
5 An exception is Gavin Whitelaw, ‘Preliminary results of a survey of Bulawayo, Shaka kaSenzangakhona’s capital from 1820 to 1827’, South African Field Archaeology, vol. 3 (1994), pp. 107-9.
6 Webb and Wright, eds., James Stuart Archive, vol. 1, p. 345.


Post-doc Thokozani Mhlambi renders a tribute to Ntsikana from his album Zulu Song Cycle.Photo courtesy of Steven Kotze.