Southwest of the Zimbabwe Plateau, where the highveld opens into the wide grasslands of South Africa’s Free State and the rolling semi-arid landscapes of Botswana’s eastern margins, the plateau’s language story continues — in a different branch, through a different path, in languages that are cousins rather than siblings of the plateau family. Sesotho. Setswana. Sepedi. The Sotho-Tswana cluster. These languages and the Zimbabwe Plateau Civilisation’s language family are not the same lineage — but they share ancient Bantu ancestry, sustained contact through the Limpopo and Kalahari corridors, and a common civilisational reach across the Southern African highlands that the colonial map deliberately divided and obscured.
The Sotho-Tswana language cluster — comprising Sesotho (Southern Sotho), Setswana, Sepedi (Northern Sotho / Sesotho sa Leboa), and several closely related varieties — is one of the largest Bantu language groups in Southern Africa by total speaker numbers, spread across South Africa, Lesotho, Botswana, Zimbabwe, and Namibia. Together these languages are spoken by an estimated 25 to 30 million people, making them collectively one of the most widely spoken Bantu language communities on the continent.
Linguistically, the Sotho-Tswana cluster belongs to the central Bantu zone — distinct from the eastern Bantu stream that produced both the plateau language family and the Nguni languages. This means the Sotho-Tswana languages and the Zimbabwe Plateau languages are related at the level of the broad Bantu expansion ancestor — cousins in the extended Bantu family — rather than through the more specific eastern Bantu ancestral lineage that connects Chikaranga to Tshivenda, Chewa, or Swahili.
In Volume 16 we established the framework for understanding distant Bantu family relationships when we examined the Chewa-Nyanja corridor — languages that are cousins of the plateau family through shared eastern Bantu ancestry but not siblings through the specific Mbire lineage. The Sotho-Tswana relationship to the plateau family is similar in structure but different in branch: the Sotho-Tswana cluster comes from the central Bantu stream, which diverged from the eastern Bantu stream at a greater depth of ancestral time than the Chewa-Nyanja divergence.
This means that the shared vocabulary between Sotho-Tswana languages and the Zimbabwe Plateau languages is real but thinner — present at the deepest proto-Bantu level, where both families share ancient roots for fundamental concepts, but absent at the more specific level of the shared Mbire civilisational vocabulary that connects Chikaranga to Tshivenda or Chikaranga to Chewa. The cousinage is genuine. It is simply more distant.
If the Zimbabwe Plateau language family and the Chewa-Nyanja corridor family are first cousins — sharing a common great-grandparent in the eastern Bantu expansion — then the Zimbabwe Plateau family and the Sotho-Tswana cluster are something like second cousins: sharing a common great-great-grandparent in the broad Bantu expansion, having diverged from each other longer ago, sharing fewer specific vocabulary items but recognisably related in fundamental grammatical structure. The family connection is there. It is simply further back in the tree.
Despite the distant family relationship, the Zimbabwe Plateau language family and the Sotho-Tswana cluster have been in sustained contact across the Limpopo and its southern approaches for centuries. This contact was most intensive in the southern plateau zone — where the Kalanga language of the western plateau sat adjacent to Setswana-speaking communities across the Botswana-Zimbabwe border, and where Tshivenda sat adjacent to Sepedi communities in the Limpopo Province of South Africa.
This contact produced specific vocabulary exchanges in both directions — plateau vocabulary entering Setswana and Sepedi through trade, political contact, and intermarriage; Sotho-Tswana vocabulary entering Kalanga and the western plateau through the same channels. The Kalanga language we examined in Volume 11 shows the most extensive Sotho-Tswana contact vocabulary of any plateau family member — reflecting its position at the western plateau margin where the two language families’ territories most directly overlapped.
The most significant vocabulary domain in which plateau language influence entered the Sotho-Tswana cluster was the gold-working and cattle management vocabulary of the Mapungubwe and Great Zimbabwe trade world. As we established in Volume 5, the Limpopo corridor was an ancient contact zone where the plateau civilisation’s gold and cattle economy interfaced with the Kalahari-edge communities to the southwest. The political and economic vocabulary of this interface — tribute language, trade protocols, certain cattle terminology — moved westward and southward from the plateau world into the Sotho-Tswana contact zone.
The stone-walling tradition also carried vocabulary: the terms for stone enclosure building, for the specific architecture of the Zimbabwe Plateau’s stone structures, appear in the traditions of communities that inhabited the southern contact zone with the plateau — including some Sotho communities whose oral traditions reference the plateau’s stone-built world.
In the opposite direction, Sotho-Tswana vocabulary entered the western plateau languages — particularly Kalanga — primarily through the ecological and economic vocabulary of the Kalahari fringe. The Kalahari Basin’s specific ecology — dryland agriculture, specific arid-zone plant and animal species, the distinctive seasonal rhythms of the semi-arid southwest — produced vocabulary that the plateau languages needed as their western communities extended into the Kalahari fringe zone. Setswana was the language already at home in that environment, and its vocabulary for the Kalahari world entered Kalanga through sustained community contact.
Despite the different lineage branches, the Sotho-Tswana languages and the Zimbabwe Plateau languages share vocabulary at the deepest proto-Bantu level — the ancestral roots that predate the divergence of the central and eastern Bantu streams. These shared roots are the most ancient linguistic evidence of their common family membership, predating any contact exchange by thousands of years.
The shared proto-Bantu vocabulary is most visible in the fundamental domains of human experience: the human body, basic kinship terms, the most ancient agricultural vocabulary, and the names of animals that have been part of the Bantu-speaking world since before the expansion southward. In these domains, the Sotho-Tswana languages and the plateau languages often show recognisably related forms — not identical, having diverged across their separate paths, but clearly related to the trained ear.
| Concept | Chikaranga | Sesotho | Setswana | Relationship |
|---|---|---|---|---|
| Person / people | munhu / vanhu | motho / batho | motho / batho | Proto-Bantu *-ntu root — all three from the same ancient source, diverged in prefix and phonology through separate paths |
| Cattle | mombe | khomo | kgomo | Proto-Bantu *-gòmbè root in all three — the same ancient cattle word diverged through separate sound change histories |
| Chief / ruler | mambo | morena | kgosi | Different political titles — the two families developed separate governance vocabulary reflecting distinct political formations |
| Ancestor / spirit | mudzimu | badimo | badimo | Related roots — both from proto-Bantu ancestor spirit vocabulary, diverged through separate paths but recognisably family |
| Land / country | nyika | naha | lefatshe | All distinct — the word for land diverged most widely, each family using a different proto-Bantu root for territory |
| House | imba | ntlo | ntlo | Sesotho and Setswana share ntlo; Chikaranga imba is distinct — reflecting the different construction traditions of each environment |
| Fire | moto | mollo | molelo | Same proto-Bantu root for fire — recognisable across all three despite divergence; one of the most widely shared Bantu roots |
| Language (prefix) | chi- / se- | se- | se- | Sotho-Tswana uses se- prefix (SeSotho, Setswana, Sepedi); plateau family uses chi-/tshi-/tji- — same noun class concept, different phonological realisation |
| Totem / clan system | mutupo / dzinza | sefate / ledimo | sereto | All three families maintain clan/totem systems with exogamy — the same ancestral Bantu social architecture expressed in different vocabulary in each family |
What the Sotho-Tswana chapter of this series makes visible — when placed alongside all the plateau, corridor, and coastal language stories we have traced — is the full scale of the Bantu-speaking civilisational landscape of Southern Africa. From the Zimbabwe Plateau to the Limpopo valley to the South African highveld to the Kalahari Basin to Lesotho’s highlands — a vast arc of Bantu-speaking communities, related at varying degrees of family distance, sharing the same ancient grammatical architecture, the same proto-Bantu roots for cattle and fire and ancestor spirits, and the same fundamental civilisational values of clan identity, governance assembly, and ancestral connection.
This is not a homogeneous space. The Sotho-Tswana family is distinct from the plateau family in language, in specific history, in ecological adaptation, in political tradition. They are not the same people. Their languages are not the same languages. But they are part of the same great Bantu civilisational landscape — related branches of a vast family tree that spread across the Southern African highlands over two thousand years of movement, settlement, and contact.
The colonial map divided this landscape into “tribes” and “ethnic groups” — each assigned a bounded territory, a single identity, a fixed location. The language evidence tells a different story: a story of sustained movement, contact, exchange, and shared family across boundaries that no colonial map could accurately contain. The plateau language family and the Sotho-Tswana cluster both know this in their vocabulary. The contact words in Kalanga and Setswana are the proof. The related proto-Bantu roots in Chikaranga and Sesotho are the proof. The shared clan exogamy system across both families is the proof.
When a Setswana speaker and a Chikaranga speaker meet, they are not meeting strangers. They are meeting distant family who took different roads from the same ancient homeland — who developed different vocabularies in different landscapes, different political traditions under different mountains, different ecological words for different skies. But the fire word is related. The cattle word is related. The ancestor word is related. That is not coincidence. That is family.
— Tete Getty, Moyo Netombo 🇿🇼The colonial map of Southern Africa imposed borders and categories on a language landscape that had been operating on its own logic for centuries. The 1885 Berlin Conference divided the Kalahari between British Bechuanaland and Southern Rhodesia — cutting through Kalanga and Setswana contact zones. The Limpopo became the border between Southern Rhodesia and the Transvaal — cutting through the Tshivenda-Sepedi contact zone and the Kalanga-Setswana overlap. The Drakensberg escarpment became the boundary around the Basutoland Protectorate — enclosing Sesotho-speaking communities in a territory defined by British administrative convenience rather than by language or cultural logic.
Each of these border-drawing acts produced the same result: language communities that had been in continuous contact, exchanging vocabulary and sharing family relationships across an open landscape, were suddenly divided into different administrative units with different legal frameworks, different educational systems, and eventually different national identities. The vocabulary exchange continued in the border communities — it always does, because language does not respect administrative lines. But the formal academic and administrative treatment of these languages stopped acknowledging the exchange, treating each language as bounded within its nation-state’s territory rather than as part of a broader language landscape that the territory’s borders cannot contain.
The colonial linguistic and administrative framework treated the Sotho-Tswana languages and the Zimbabwe Plateau languages as belonging to separate, unrelated “ethnic” spheres — drawing hard distinctions where the language evidence shows a gradual, contact-shaped continuum. The vocabulary exchange between Kalanga and Setswana was ignored. The Tshivenda-Sepedi contact zone was underanalysed. The shared proto-Bantu roots connecting the two families were treated as irrelevant to the administrative categories of the nation-states they happened to sit within. The language families of the Southern African highlands form a civilisational continuum, not a collection of bounded ethnic units. The vocabulary exchanges across their contact zones are the proof. This series names that continuum and refuses the administrative fiction that separated it.
From the Kalahari to the Limpopo Province to the Free State highlands, the plateau language family’s southwestern reach is still audible in the contact vocabulary of living communities.
With this volume, the language map of this series is complete. We have traced the Zimbabwe Plateau Civilisation’s language family in every direction: southward through the Limpopo to Tshivenda (Volume 8), northward through the Zambezi to the Tonga and Korekore (Volumes 12–13), eastward to the Indian Ocean through Ndau and the trade routes (Volume 15), northeastward to the Malawi corridor through Chewa and Nyanja (Volume 16), and now southwestward through the Kalahari fringe and the Limpopo contact zones to the Sotho-Tswana family’s distant cousinhood.
The plateau did not exist in isolation. It was the heart of a vast Southern African language landscape — connected by corridors and contact zones to language families in every direction, enriched by trade vocabulary from the Indian Ocean world, related by ancient family to the Sotho-Tswana communities to its southwest and the Chewa-Nyanja communities to its north. It was a civilisation in conversation with its continent — and the vocabulary of that conversation is still readable in the living languages of the region today.
In the final volume of this series — Volume 24 — we step back from the map and ask the larger question: what does it mean to dismantle the language boxes entirely, and what would a truly civilisational linguistics for Southern Africa look like? The 5,000-year arc is complete. The restoration remains ongoing.
Re bua leleme le le lengwe la bogologolo. Le fa re bua ka dipuo tse di farologaneng, re tswa molapo o le mongwe. We speak different ancient tongues. Yet we come from one river.
Leave a Reply