Skip to content
Language Lineage

Proto-Bantu

The reconstructed ancestor of about 500 languages, built around a system of noun classes marked by prefixes: *mʊ-ntʊ 'person', *ba-ntʊ 'people', the plural that gave the family its name.

c. 3000 BCE (reconstructed)The Grassfields of the Nigeria–Cameroon border
On this page
  1. Origins
  2. The Expansion
  3. Where its languages are spoken
  4. Turning points
  5. Contested classifications

01

Origins

Every Bantu language sorts its nouns into classes, each marked by a prefix that also appears on adjectives, verbs and pronouns that agree with it. The classes come in singular–plural pairs: Proto-Bantu *mʊ-ntʊ 'person' has the plural *ba-ntʊ 'people', and in 1862 the linguist Wilhelm Bleek took that plural as the name of the whole family. Swahili mtu/watu, Zulu umuntu/abantu and Kikongo muntu/bantu still show the pattern.

The homeland is placed where the related, non-Bantu Bantoid languages are most diverse: the Grassfields on the border of Nigeria and Cameroon. Proto-Bantu vocabulary includes words for yams, oil palms, goats, pottery and fishing, but its speakers did not yet work iron.

02

The Expansion

Over roughly three thousand years, Bantu-speaking farmers spread across most of sub-Saharan Africa south of the equator, one of the largest language expansions in history. The spread happened in stages. Whether the ancestors of the Eastern Bantu left early, skirting the northern edge of the rainforest, or late, through it, has long been debated. Phylogenetic work by Grollemund and colleagues (2015) favours a route through the forest, opened when a drier climate around 500 BCE broke up the rainforest into savanna corridors.

This tree follows the main lines of those studies: Northwest Bantu, the languages that stayed nearest the homeland, Western Bantu in the lower Congo and Angola, and Eastern Bantu, which spread from the Great Lakes to the Cape. Malcolm Guthrie's letter-and-number zones (A10, G42 and so on) are still used as reference codes, but they are geographical, not genealogical.

On the map

Where its languages are spoken

Comoros
Spoken todayShading marks whole countries, not the regions within them.
  1. South AfricaKusi c. 49 million
  2. RwandaGreat Lakes Bantu c. 13 million
  3. ZimbabweNyasa · Kusi c. 13 million
  4. AngolaWestern Bantu c. 12 million · Luban
  5. BurundiGreat Lakes Bantu c. 12 million
  6. BotswanaKusi c. 2 million
  7. LesothoKusi c. 2 million
  8. EswatiniKusi c. 1 million
13 more countries
  1. ComorosSabaki c. 800,000
  2. CameroonNorthwest Bantu
  3. Democratic Republic of the CongoWestern Bantu · Luban · Great Lakes Bantu
  4. Equatorial GuineaNorthwest Bantu
  5. GabonNorthwest Bantu
  6. KenyaSabaki
  7. MalawiNyasa
  8. MayotteSabaki
  9. MozambiqueNyasa · Kusi
  10. Republic of the CongoNorthwest Bantu · Western Bantu
  11. TanzaniaGreat Lakes Bantu · Sabaki
  12. UgandaGreat Lakes Bantu
  13. ZambiaLuban · Nyasa

Figures are rough first-language estimates, rounded, from censuses and Ethnologue. Counts for small or scattered communities are especially uncertain.

Rings along the limb

Turning points

Oldest first.

  1. Splitc. 3000 BCEContested

    Leaving the Grassfields

    Proto-Bantu speakers begin to move south and east from the Nigeria–Cameroon borderlands into the forest of southern Cameroon and Gabon. The date is estimated from linguistic dating and archaeology and is approximate.

    Sources (1)
    • Grollemund, Rebecca, et al. 2015. Bantu expansion shows that habitat alters the route and pace of human dispersals. Proceedings of the National Academy of Sciences 112(43).
  2. Splitc. 1000 – 500 BCEContested

    Through the rainforest

    As the climate dries and savanna corridors open in the Central African rainforest, Bantu speakers spread through it to the south and east. The 'late split through the forest' now has more support than the older 'early split around the forest', but the question is still argued.

    Sources (2)
    • Bostoen, Koen, et al. 2015. Middle to Late Holocene paleoclimatic change and the early Bantu expansion in the rain forests of Western Central Africa. Current Anthropology 56(3).
    • Grollemund, Rebecca, et al. 2015. Bantu expansion shows that habitat alters the route and pace of human dispersals. Proceedings of the National Academy of Sciences 112(43).

Outside the tree

Contested classifications

These disagreements may never be settled. Each position is given with its grounds, and no side is declared correct.

  • Widely accepted

    Niger-Congo

    Bantu's wider family, Atlantic-Congo, is related to Mande, Dogon, Ijoid and other West African groups in one macro-family

    Evidence for

    Greenberg's classification (1963) grouped these languages on shared noun-class prefixes and vocabulary, and Niger-Congo is still the conventional label in handbooks and surveys. Bantu's place inside Benue-Congo and Atlantic-Congo is well established.

    Evidence against

    Güldemann, Dimmendaal and others argue that the outer members, especially Mande and Dogon, lack the noun-class system and have never been shown to share regular sound correspondences with Atlantic-Congo. On this view 'Niger-Congo' is a hypothesis, not a demonstrated family, and Glottolog treats its outer groups as separate families.

    Sources (2)
    • Greenberg, Joseph H. 1963. The Languages of Africa. Bloomington: Indiana University.
    • Güldemann, Tom. 2018. Historical linguistics and genealogical language classification in Africa. In The Languages and Linguistics of Africa, ed. Tom Güldemann. Berlin: De Gruyter Mouton.