Iron Age studies in South Africa are dominated by Huffman’s (1982, 1986, 1993, 2001) ethnographically derived Central Cattle Pattern model, which identifies the cattle-based bridewealth institution of South Eastern Bantu-language speakers by the spatial distribution of specific archaeological features. The idea of the spatial expression ‘on the ground’ of a variety of symbolic codes was Adam Kuper’s (1980, 1982) interpretation of predominantly Swazi ethnography. Surprisingly, Kuper’s work has never been interrogated and consequently his misunderstanding of the ethnography was carried into the Central Cattle Pattern and interpretations of the last 1600 years of Iron Age, farmer archaeology in southern Africa. Two particular features, burials and subterranean grain storage pits, and their relationship to cattle-kraals are explored. Because cattle are central to the Central Cattle Pattern, much archaeological attention has been given to looking for evidence of cattle at archaeological sites, either by dung, bones or cattle-kraals. The paper presents the views of contemporary Swazi, Xhosa and Mfengu people that suggest the symbolic importance of cattle-kraals; in the extreme they may not reflect the presence of livestock at all, yet their persisting presence demonstrates the continuing importance of cattle, real or imagined.