All events · Conquest and the Slave Trades
The Battle of Mbwila, the Kongo Civil Wars and the Emptying of the Kingdom
c. 1665–1740 · Sub-Saharan Africa
Displacement
Territory Seized
Displaced
200,000
No census exists. The scale rests on the trans-Atlantic slave-trade voyage database, which records roughly 500,000-600,000 people embarked from the Loango-Congo coast between 1660 and 1740, and on John Thornton's reconstruction of the civil wars from Kongo, Portuguese and Capuchin correspondence; Thornton argues a large share of those embarked in this period were prisoners of the Kongo wars. The figure of about 200,000 is a conservative middle estimate of Kongo-origin captives, not a count.
Share of the population affected
40%
of the population of the Kingdom of Kongo on the eve of the Battle of Mbwila, 1665
Open on the interactive map →
What happened
At Mbwila (Ulanga) in October 1665 a Portuguese-Angolan force with African allies defeated and killed King Antonio I of Kongo, an outcome that removed the kingdom's central authority. Rival branches of the royal family then fought for the throne for two generations; the capital Mbanza Kongo was abandoned in 1678 and only reoccupied under Pedro IV in 1709, after the Antonian movement of Beatriz Kimpa Vita was suppressed and its leader burned in 1706. Prisoners taken in these wars were sold to Portuguese and Dutch traders at Luanda, Cabinda and Soyo, and the region became one of the largest sources of captives in the Atlantic trade. Historians differ on whether the wars drove the slave trade or the trade drove the wars: John Thornton emphasises internal dynastic politics with slaving as a consequence, while other scholars stress European demand as the motor. Kongo survived as a title and a set of provinces but never recovered its former reach.
Parties
Portuguese Angola and rival Kongo royal houses fought wars that broke up the kingdom and fed captives into the Atlantic trade
Disputed
Whether the civil wars generated the slave exports or were generated by them, and what share of Loango-Congo embarkations were Kongo war captives, remain matters of scholarly disagreement.
The denominator
John Thornton's reconstruction, built from Capuchin mission baptismal registers, provincial tribute lists and territorial extent, puts the kingdom at roughly half a million in the mid-seventeenth century. It is the most carefully evidenced population estimate for any pre-colonial Central African state, but it still rests on extrapolation from mission districts to the kingdom as a whole.
The 200,000 accumulates across 75 years of intermittent civil war, so it spans roughly three generations while the denominator is a single-year estimate.
Where the baseline comes from
On the death toll
No usable death toll exists. Contemporary Portuguese accounts of Mbwila claim thousands of Kongo dead including the king and much of the nobility, and the later wars were fought largely as raids on villages; deaths in captivity and on the Atlantic crossing are counted in the slave-trade entries rather than here.
Documented forms of violence
Forced LabourDestruction of Property
Capuchin missionary reports and Portuguese colonial correspondence describe villages burned and their inhabitants enslaved during the succession wars.
Movements mapped
- Kongo captives to Luanda and the Atlantic ships
Countries touched, on present-day borders
Angola
Related entries
- Overlaps with: Portuguese Angola and the West Central African Slaving Wars
The Angola entry covers Mbwila and the Kongo civil wars as part of West Central Africa's slaving wars, so the two entries describe overlapping captives.
- Part of: The Trans-Atlantic Slave Trade to 1750
The Kongo entry's ~200,000 is explicitly an estimate of Kongo-origin captives within the 500,000-600,000 embarked from the Loango-Congo coast in 1660-1740, taken from the same slave-voyages database. All of these embarkations fall before 1750, so they are inside the trans-Atlantic entry's 4.8 million. Kongo's internal displacement (the abandonment of Mbanza Kongo, 1678-1709) is not quantified and is not shared. Deaths are not shared: the Kongo entry has none, and its note already assigns deaths in captivity and at sea to the slave-trade entries. 200,000 people are counted in both entries.
Sources