How Societies Are Born: Governance in West Central Africa before 1600

4. Of Courts and Titleholders

4     E        population density in West Central Africa discloses a striking contrast between two nearly adjacent large areas of higher population density inland from the coast in western middle Angola, on the one hand,and all the surrounding regions on the other (map ).In addition, there exists a set of smaller population clusters strung from east to west around latitude ° to ° S and another one in the floodplains of the upper Zambezi. The two nearly adjacent large population clusters are located on the Benguela planalto, now inhabited by Umbundu speakers, and in the middle Cuanza basin and its environs, now settled by Kimbundu speakers. Neither of these clusters are the product of recent colonial situations. The one in the Cuanza region was already noted around  while the existence of the other was reported by the later eighteenth century and documented in more detail around .1 It could be argued that these clusters resulted from population movements caused by refugees fleeing Portuguese wars of conquest and slave raiding in the seventeenth and eighteenth centuries. Indeed, considerable population transfers did occur in the Cuanza Valley, especially between ca.  and  when the lands between Massangano and the coast were partly depopulated, while those north of Ambaca saw an influx of refugees.2 Yet in the wider context even such sizeable movements only slightly displaced the center of gravity of an earlier cluster of high population density just as that center would later gravitate to the lands around Malange after . . Filippo Pigafetta and Duarte Lopes, Description du Royaume, p. , concerning the Kingdom of Ndongo between Cuanza and Lucala, comments,“very dense population , more than one would believe,”and p.  speaks of “a dense population”in the lands of the middle Kwilu basin. For the planalto, see Heywood and Thornton, “African Fiscal Systems,” pp. ‒ (map of reconstructed population density: p. ). Data from the Cuanza basin from ca.  to ca.  allow population densities to be reconstructed there, but this has not yet been done. . Beatrix Heintze,“Gefährdetes Asyl,” pp. ‒.      Given this case, it is reasonable to assume that something similar occurred on the planalto. Joseph Miller has argued that many people fled from the Cuanza Valley and the Portuguese dominions to the northern and western rims of the planalto starting in the later seventeenth century.3 But was this displacement substantial enough to produce a dense population over the entire planalto? Quite substantial numbers of undated and mostly unexplored stone ruins of walls and tombs everywhere on planalto seem to attest to the existence of a population cluster, but then again some have . Joseph Calder Miller, Way of Death, pp. , ‒, summarizes many sources, including Cavazzi. Oral traditions about the dynastic origins of most of the western and northwestern Ovimbundu kingdoms, especially Bailundu, but also some of the kingdoms on the southwestern planalto refer to late–seventeenth-century origins in the Cuanza basin. + + + + + + + + + Etosha Pan Okavango Okavango Swamp Swamp Okavango Swamp O k a v a n g o R . ATLANTIC OCEAN Cunen e R . Cuando R . Z a m bezi R. L u a l a b a R . L u b i l a s h R . K a s a i R . L u l u a R . K w i l u R K w a n g o R . C o n go R. Loge R. C u a n z a R . Qu e v e R . 25o E 20o E 15o E 15 o S 10o S 5 o S 4 o S Cuando: Rivers Etosha Pan: Swamps, Pans 0 300 km Over 10 people per km 2 in 1965 Ancient population cluster Ancient population density very low +  . Population Clusters (‒) claimed that all of these ruins also date only from the late seventeenth century onward (map ).4 Although these claims are certainly exaggerated, only further archaeological research can settle the chronological issues. Meanwhile the population density map still suggests that clustering first took place somewhere north of the middle Cuanza basin and also somewhere on the      . See Mesquitela de Lima, ‒, but he himself believes that most walls belong to sites dating before the fifteenth century. Most graves and cemeteries may well date to the eighteenth and nineteenth centuries, but some definitely existed in the seventeenth century (Manuel Gutierrez, Archéologie) and an illustration in Cavazzi (António de Montecúccolo Cavazzi, Descrição, :book , par. , ). Others may well be older. Beatrix Heintze gave a state...