Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, African History. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 06 July 2022

Kingdoms of South-Central Africa: Sources, Historiography, and Historyfree

Kingdoms of South-Central Africa: Sources, Historiography, and Historyfree

  • David M. GordonDavid M. GordonDepartment of History, Bowdoin College


In his influential book, Kingdoms of the Savanna (1966), Jan Vansina described the rise of the kingdoms of the south-central African interior from the 15th century. These include the Luba (the mulopwe titleholders), Lunda (the nuclear Lunda, also termed Rund, of the mwant yav titleholders), Lunda-Ndembu, Chokwe, Pende, Luvale, Luluwa, Kanyok, Luba-Kasai, Kuba, Eastern Lunda, Yeke, and the Bemba. New analyses of oral traditions as well as the study of art, archaeology, ethnographic fieldwork, linguistics, and documentary sources haverevised understandings of these polities and added details. Historians have considered the context of the production of primary sources, in particular art and oral traditions, which were created during a transformative 19th century, when trade and violence contributed to the centralization of power for some polities and the disintegration of others. With subjects questioning the power of sovereigns, art, oral traditions, and oral praises projected royal genealogies and the qualities of kingship into a vague antiquity. The study of historical linguistics has also provided inroads into understanding the dissemination of political institutions and titles along with tentative accounts of their historical depth. Ethnographic fieldwork has further elaborated on the functioning of political systems and religious ideas. These diverse primary sources complicate the historiography of central African kingdoms; they also indicate the spread of alternative political and religious affiliations during the 18th and 19th centuries, in particular Luba fertility associations and Lunda fictive kin alliances.


  • Central Africa
  • Historiography and Methods
  • Oral Traditions

As I set up my camera I was told to wait a bit and two men came in carrying a flat wooden case. This was carefully laid upon the ground and opened. It was a large standing mirror. The glass was adjusted so that the King could see himself and he ordered a change in the set of his crown. The King was finally satisfied and I went to work. I shot him with everything I had . . . with the Li[n]hof in black and white and color, with the Rolleiflex and Contax. I threw the works at him.

— Life photographer Eliot Elisofon on photographing the Kuba king1

Central African royals and their entourages have constructed or encouraged the construction of paraphernalia of power: imaginative myths, histories, praises, art objects, and emblems. Further celebrated in photography and in the written record, these constructions form the basis of professional historical accounts of south-central African kingship. The historiography of south-central African kingship is best addressed by pointing to the evidence upon which it rests. The imaginative accouterments of kingship that underpin historical knowledge inform the central line of discussion of this article.

Figure 1. Mbop Mabiinc maMbeky (ruled 1939 to 1969), Kuba King, Nsheng, Belgian Congo. Photograph by Eliot Elisofon, 1947. Eliot Elisofon Photographic Archives.

The savanna woodlands of south-central Africa are cut through by the northward flowing tributaries of the Congo River and southward flowing tributaries of the Zambezi River. Fish-rich lakes and fertile flood plains. amidst a terrain of poor soils and unevenly distributed natural resources contributed to sparse and mobile populations of “paddlers and pedestrians,” as Thomas Reefe puts it in his masterful synthesis of the early modern political history of this region.2 Decentralization and localization was the norm here, with longue durée forms of associational life organized around these characteristics. Yet in the region between the dry Kalahari scrublands and the equatorial forest of the Congo River basin, in the Miombo woodland savanna, located in the most fertile and resource-abundant areas, forms of political centralization and organized religion emerged. As regional and international trade in slaves, cloth, iron, guns, copper and other luxury goods accelerated through the 18th century, affiliations became even more extensive. Individuals mobilized power and set up hereditary political offices or titles. These are the “kingdoms” that have become a source of fascination for central Africans and their historians.

Which political systems qualify as kingdoms? At a minimum, a kingdom must mean that a hereditary titular office has become more powerful than other political offices and a defining political characteristic. The inheritance of a political office of a powerful title is a low bar, and many central African polities meet this criterion; on the other hand, if we define a kingdom as an absolute and indivisible sovereignty, along the lines of medieval and early modern Europe, fewer examples of central African kingship can be identified. At what point should the historian decide that autonomous and decentralized chiefdoms give way to centralized kingdoms?3 And does it matter how far back in time such inherited titles emerged? Some central African kingdoms only became hereditary offices in colonial times, and hence sovereignty has never actually resided in a king but rather in the modern colonial and postcolonial state. Does a kingdom have to precede a modern state, however? The question remains: beyond the recognition of an inherited title of those who claim significant power, what is meant by a “kingdom”?

The signature kingdoms of south-central Africa, the Luba and Lunda, which are the core examples here, were in fact several different polities tied together by changing hierarchical and horizontal affiliations. There can be many types of kingdoms; there is no reason why Luba and Lunda polities should not be worthy of the label. The challenge is to ensure that the term “kingdom” does not homogenize different alliances, political hierarchies, and religious associations. European examples of medieval and early modern kingdoms should not be used as models. Once a word that evolved in the European political context such as “kingdom” is employed, the historical evidence of political affiliations tends to be selected and interpreted in a particular fashion. Claims for African kingdoms along European lines often draw out their historical arguments by cherry-picking the most select fruit.

When considering the entire region of south-central Africa from the usual starting point of the rise of the kingdoms (the 16th or even 17th century) much generalization is necessary. The temptation for an encyclopedic synthesis is to develop general historical patterns, as has been done on several occasions, beginning with the influential Kingdoms of the Savanna, by Jan Vansina, and more recently by Thomas Reefe and Isidore Ndaywel è Nziem.4 Instead, dwelling on historical evidence and its interpretation, it is helpful to consider the evolution of representations through which kingship attained a certain historical and historiographic reality. This article takes Vansina’s observation of the Luba as “first and foremost a construction of the mind” as a starting point.5 How has this construction of the mind been represented in documents, art and oral tradition, performative rituals, political titles, and words? Which aspects of kingship are revealed through these imaginative constructions? Then, how have historians rendered them into a historiography of central African kingdoms? The idea here is to open up a historical discussion about central African kingdoms rather than providing an authoritative account.

The key examples considered here include polities established around the southern tributaries of the Congo River, that is 21st-century Kasai and Katanga provinces in the DRC, northern Zambia, and western Angola. They include the Luba (in particular the Luba-Katanga of the mulopwe titleholders), Lunda (the nuclear Lunda, also termed Rund, of the mwant yav titleholders), Lunda-Ndembu, Chokwe, Pende, Luvale, Luluwa, Kanyok, Luba-Kasai, Kuba, Eastern Lunda, Yeke, and Bemba. Examples from these polities, along with key literature, will be included here rather than attempting to provide comprehensive histories of each polity. Some southern-oriented polities—such as the Lozi and Ngoni—are notably absent. Likewise, the western polities of the Kongo, Ndongo, Imbangala, and Ovimbundu, are not considered here. Even as they shared certain attributes with the Luba and Lunda, there are significant historical divergences including different constellations of trade and conquest, notably the Zulu consolidation to the south and closer proximity to Portuguese and Brazilian slaving to the west.

Map 1. Kingdoms and Polities of the South-Central African Interior, c. 18th–19th Century.

Scholars place African kingdoms between poles of violent exploitation and consensual leadership. An initial phase in Africanist historiography finds kingdoms emerging from a fascination with power almost autonomous from material interest and underlying modes of production. Influenced by the notion of divine kingship, this historiography attained great sophistication in Jan Vansina’s early work along with that of several anthropologists and art historians.6 A complementary view emphasizes the materialist consensual arrangements that led to the centralization of political power, such as the need for collective effort for agricultural or hunting endeavors (or both), contributing to a division of labor and a hierarchical consolidation of rank and status. On the other hand, a conflictual approach, typical of the historiography of those west African kingdoms empowered by the Atlantic and Saharan slave trades, focuses on military force and exploitation. Here the artifacts of power—symbols, narratives, and religion—are viewed as legitimizing devices rather than consensual constructions.7 Such historiographic interpretations need not be mutually exclusive: a set of affiliations based on consent could easily become one of force. Moreover, the perspectives of those supporting or of those opposing a kingdom might differ; not all subjects, followers, captives, and slaves of central African kings sang the same praises with the same enthusiasm.

Central African kingship is entwined with questions at the heart of Africanist historiography. In addition to methodological innovations in the use of oral traditions (along with linguistics, ethnography, and archaeology), the historiography of African kingship intended to demonstrate that Africans developed complex political institutions, combating racist notions that viewed political complexity inspired by outside influences, such as the dissemination of “Hamitic” civilization or international traders. To demonstrate that these were African achievements and not those of foreign agents, scholars pushed back the date of the founding of these kingdoms. This article is in many ways an outcome of that historical scholarship, yet it is not as concerned with proving that such agency occurred in the absence of global forces. It points to forms of regional and international interconnectedness that contributed to political change. Nor is kingship viewed here as a more complex or advanced achievement than other, sometimes decentralized, political arrangements. Thus, even as the focus here is on kingship, there are attempts to look beyond kingship as a key political, social, and religious development.

There are four broad and overlapping evidentiary pathways that develop distinctive narratives of kingship even as they influence one another. Each form of evidence is considered here in turn: oral tradition, art and material culture, fieldwork-based ethnography and linguistics, and documentary evidence. A final section develops tentative conclusions about political and religious sovereignties from the 16th to 19th centuries.

Oral Traditions

The historiography of central African kingdoms is rooted in the study of oral traditions. The most popular, widespread, and influential are stories of origin, exciting and passionate narratives, performed frequently and written about fairly often. Over the last five hundred years, such “traditions of genesis” spread across the savanna as different groups and lineages adopted them. By doing so, rulers legitimized their authority and claimed far-flung affiliations. Since political titles appear in the traditions as narrative characters, the traditions have a palimpsest-like quality and are thus evidence of far more than the genesis of kingdoms. Historical events and processes that occurred hundreds of years apart may appear within single narrative episodes. A political title that is a character in the narrative could be the composite of the lives of several individuals who occupied that title.

In addition to narratives, oral traditions also include praises of distinctive kings, heroes, and other titleholders that record their traits, accomplishments, and characters. Even vernacular intellectuals close to the royal courts have lively debates over the meanings of these poetic and sometimes esoteric praises. Nonetheless, they provide historians with some of the qualities of leadership and help to distinguish individuals from the titles they occupy.

Lineage elders or groups who claimed ownership of the narratives remember and perform oral traditions. Thus, these performed narratives were not created and developed as histories to be universally appreciated. For example, a Luba subject would not recite a history that did not belong to him or her. Sometimes those storytellers who owned the history were initiated into secretive societies, as in the men of memory of the Luba bambudye society, helped by the use of mnemonic devices such as the lukasa boards (see Figure 2). Through color and pattern association these boards evoked aspects of the oral tradition, tied it to titleholders, and located it within a physical landscape. Lukasa boards, together with the oral tradition, thus acted as a sort of historical map that could only be read by expert storytellers.8

Figure 2. Luba Lukasa Memory Board. Late 19th or early 20th century. Dimensions 25.4 x14.6 x4.4 cm. Brooklyn Museum, Gift of Marcia and John Friede, 76.20.4. Creative Commons License.

Scholars transformed these insider narratives into histories to be generally appreciated: that is, universal histories. The most effective narratives to work with were the broadly shared traditions of genesis. Historians detected the historical origins of kingdoms in such shared myths; anthropologists explored the cosmological and ideological basis of kingship through them.

The popular Luba tradition begins with a cruel and uncivilized despot, Nkongolo (meaning “rainbow”), born of the coupling of two serpents in the sky. He has the qualities of redness and cruelty. A wandering hunter, Mbidi Kiluwe (“Mbidi, the Hunter”), encounters two of Nkongolo’s sisters, Mabela “the liar” and Bulanda “sadness.” Mbidi, with his black skin and filed teeth, is handsome and demonstrates his superior behavior. Upon his presentation to Nkongolo’s court, Mbidi so impresses Nkongolo that he gives his two sisters to Mbidi, who impregnates them before he departs once again. Bulanda gives birth to Kalala Ilunga (Kalala, the warrior) who, when grown, threatens the rule of Nkongolo. After Nkongolo attempts to assassinate Kalala, he flees, gathers an army from the realm of his father’s people, and defeats Nkongolo. Kalala Ilunga establishes the Luba kingdom and becomes its first king, the mulopwe.

These are but the bare narrative outlines of a story that varies in details according to narrator. Using a version recorded in 1973 from a man of memory belonging to the bambudye society, historian Thomas Reefe identifies twenty-six distinct episodes in the tradition of genesis.9 Reefe finds these episodes occasionally and sometimes switched in seventeen other versions of the oral tradition recorded and transcribed since the beginning of the 20th century to the 1970s. The most thorough version, gauging by Reefe’s twenty-six episodes, was recorded by the missionary Harold Womersley.10 From the geographic spread and variations of the genesis myth, the story is probably of some antiquity. It is unclear whether the story emerged at the center of a kingdom and then disseminated, or was itself an amalgamation of even older myths from across the region (it was probably a combination of both). Were the characters in the narrative actual historical individuals or mythical idealizations? Even if they existed as individuals, they are clearly the product of a vibrant cultural imagination that invested them with idealized and mythical qualities.

At first, to give the oral traditions legitimacy as evidence, historians reduced the genesis myths to the key narrative elements they considered plausible historical events. In his influential account published in 1966, Kingdoms of the Savanna, Vansina pared down the oral tradition to the conflict between historical personalities of Nkongolo and his nephew, Kalala Ilunga, identifying them as the founders of the first and second Luba kingdom (and ultimately “empire”) respectively.11 Vansina would later insist on the palimpsest quality of such oral traditions and question several aspects of his Kingdoms.12 Still, historians often treat the Luba characters in these myths as historical founders. Congolese historians, in particular, inspired by Vansina’s initial account and their proximity to reconstruction of these traditions as aspects of their national traditions, remain insistent on the historical veracity of the characters of the oral tradition. Most notable is Isidore Ndaywel è Nziem’s thorough attempt to reconcile various renderings of the oral traditions and come up with definite locations for the origin myth. Confirming aspects of Vansina’s initial work, Ndaywel è Nziem agrees that the oral tradition dates to at least the 16th century. (As indicated below, however, based on archaeological evidence, Ndaywel è Nziem provides an earlier date of c. 13th century for the origins of Luba civilization.)

Luc de Heusch pioneered the study of central African oral traditions as myths, less concerned with the historical veracity than with the cosmology and intellectual universe of kingship. De Heusch finds these the Luba myths to be a “latent cosmic drama that the Luba disguise as history.”13 Nkongolo, the evil despot, symbolizes infertility and incest; Mbidi, the sacred civilizer, brings fecundity and exogamy. The myth introduces a linear history that overcomes the cyclical nature of endogamy and repetition that led to sterility and drought. Nkongolo’s world is of incest; the first exogamous marriage occurs with the coming of Mbidi, the hunter outsider whose proximity to nature is a source of fecundity. He breaks the pattern of incest and provides the cultural revolution that underpins Luba kingship. Mbidi Kiluwe and Kalala Ilunga still inherit the residue of this incestuous world; the oral tradition represents the continued tensions between patrilinearity—ties between fathers and sons (Mbidi Kiluwe and Kalala Ilunga)—and matrilinearity—ties between uncles and nephews (Nkongolo and Kalala Ilunga). The stranger, the father Mbidi Kiluwe, who overcomes incest and sterility, promises fertility and fecundity.

In the most significant single study of the Luba-Katanga polity, Thomas Reefe, accepting some (but by no means all) of the symbolic interpretations of de Heusch, treats the myth as political charter rather than a founding history. Reefe argues that the tradition links titles, land, and places to Luba expansion. Most importantly, it affirmed the matrilineal succession of some titles among the Luba but also demonstrated that a male from the patriline would eligible for the throne: that is, the title of mulopwe, glossed by Reefe as the “sacral king” of the Luba. These mulopwe were descendants of Kalala Ilunga, distinct from mere men of power and opportunity such as the mukalanga and thus the original Nkongolo. “The struggle, then, was really between a royal male and the maternal uncles of his rivals, and in the genesis myth Kalala Ilunga is the royal male who will rule while Nkongolo is the symbol of the rival maternal uncle.”14 In addition to explaining succession, Reefe argues that the genesis myth elevated Luba mulopwe to the status of ancestral deities, who themselves became tied to the landscape. Nkongolo’s head and genitals were buried in his own sacred village, as would become the practice for subsequent chiefs. The village became the residence of female spirit medium of these ancestral chiefs, the mwadi, and attained representation in the oral tradition, on the lukasa boards, and through their sacred graves on the landscape itself.15

A second major tradition of genesis is celebrated by the Lunda, a range of polities in eastern Angola, northwestern Zambia, and southwestern DRC. At the core of this group of affiliations, once termed a Lunda empire but better described as a loose set of affiliations (a Lunda commonwealth) is the so-called nuclear Lunda ruled by the mwant yav titleholder, referred to here as the Rund (Rund henceforth refers to the particular nuclear Lunda polity and Lunda to the general set of groups who affiliate themselves to a common oral tradition of genesis). The Lunda oral tradition reflects similar contestations in history and in scholarship as the Luba. Among the Lunda affiliates, and especially the Rund, political institutions such as perpetual kinship and positional succession (see the section on Ethnographic Fieldwork and Historical Linguistics) transformed and telescoped aspects of the oral tradition. Moreover, documentary accounts of local titles, linked to the Luso-African trade from Angola, have complicated the chronology established by the oral tradition.

The Lunda oral tradition appears to be connected to the Luba. Its begins with a drunken dispute between a father and his two sons, Kinguri (or Chinguri) and Kinyama (or Chinyama). As a result of the dispute, the father disinherits his sons and gives the symbol of rule, the lukano (rukan), an iron bracelet wrapped with human tissue, to his daughter, Luij (or Ruwej). A Luba prince and the grandson of the Luba hero referred to above, Mbidi Kiluwe, “Chibind” Yirung (Yirung from the Luba, Ilunga), finds Luij. Luij marries Chibind and gives the lukano bracelet to him, signifying his rule. She proves incapable of bearing a son, so Luij allows Chibind Yirung to marry a junior wife who has a son, Yav a Yirung (Ilunga). Following the assassination of Chibind Yirung, this son becomes the mwant yav (Lord Yav), the first in the Rund kinglist. Angry with their dispossession, Luij’s brothers, Kinguri and Kinyama, leave the Rund to establish their own kingdoms. Kinguri becomes an Imbangala titleholder in Kasanje near Portuguese Angola, and Kinyama becomes a titleholder among the Luvale, south of the Rund.16

In Kingdoms of the Savanna, Vansina accepted the affiliation between the Luba and the Lunda indicated in the oral tradition, allowing for the possibility that the Lunda hunter-hero, Chibind Yirung, was the grandson of the Luba hero Mbidi Kiluwe. At the very least for Vansina, the tradition indicated Luba conquest of Lunda lands or the introduction of Luba titles into Lunda areas, or both. “The crucial event in the earlier history of Central Africa,” Vansina writes in Kingdoms, “has been not the creation of a Luba Kingdom by [N]Kongolo and Kalala Ilunga, but the introduction of Luba principles of government into Lunda land under Cibinda Ilunga [Chibind Yirung] and their transformation by the Lunda.”17 Given the initial estimation of the founding of the Nkongolo-Mbidi Kiluwe encounter to be in the 16th century, and the hypothesis that Chibind Yirung was the real grandson of these characters, Vansina dated the Lunda emergence to the early 17th century and the Luba approximately a century before. Ndaywel è Nziem accepts this dating but once again proposes to extend it to an earlier period, around the 14th century, based on archaeological evidence (see the discussion on Art and Material Culture).18

While the Lunda oral tradition may not have had the same symbolic focus as the Luba (in particular the concern with fertility), it also requires careful interpretation based on an understanding of Lunda political culture.19 Anthropologist Manuela Palmeirim, for one, considers the oral tradition as a “constellation of individual narratives, melded together, with different emphases depending on the performer and performance.”20 Such interpretations have provoked historians to question literal interpretations of the tradition, in particular the addition of peripheral characters to the core tradition. These characters had once served to date narratives, since they can be identified in the documentary record. However, it is not clear that the characters mentioned in the oral tradition were present in early renditions of the tradition. Revising the point at which key characters were added to the tradition can change the dating of the founding—or at least expansion of the Rund and the creation of the broader Lunda commonwealth.

The most important revision that has significant historiographic implications is that Kinguri, the brother of Ruwej, was a later addition to the Lunda tradition, and not contemporaneous to the founding of the kingdom.21 This revision de-anchors the dating of the Rund kingdom and throws into question much of the early history of the Lunda expansion, as recorded in Kingdoms.22 Kinguri, in addition to being Luij’s brother, was a widely known hero across west-central Africa and can be dated as an Imbangala titleholder at the court of Kasanje in a Portuguese account dating to 1655–1660 (perhaps referring to events fifty years prior). Kinguri’s exodus had thus provided a historical anchor for the dating of the Rund genesis and expansion toward Angola.

Joseph Miller first questioned the periodization of Kinguri’s exodus. He pointed out that Kinguri was an Imbangala title, not an individual, and that the other characters in the Lunda tradition could have also been perpetuated titles. This suggests that the conflict described in the oral tradition may have been a telescoping of struggles, real or metaphoric, between different titleholders.23 Initially Miller claimed an older date for the Lunda state; however, the effect was also to further free historians from a literal reading of the oral tradition. Around the same time Jean-Luc Vellut, considering the kinglists, suggested a later dating of Rund origins, which did not match the ostensible departure of Kinguri based on the Portuguese documentary record.24 In a 1998 article, Vansina argued that Lunda expansion to the south and west only occurred in the 18th century and probably reflected the dispersal of Lunda titles and Lunda stories to the west and south.25 Titleholders from the west such as Kinguri were in turn incorporated into Lunda oral traditions in order to confirm alliances manifest in the caravan trade of the 19th century.26 The same analysis could apply to Kinyama, the other brother of the Rund heroine, Luij, who became a title in the Upper Zambezi Lunda-related polities—in particular the Luvale.27 “Once one rejects the traditional evidence about Kinguri’s companions,” Vansina concludes, “there is no evidence that any of the many chiefdoms in this vast area were either founded or more modestly taken over from autochthonous chiefs by Rund migrants. There was never any Lunda [Rund] expansion to the south.”28

Instead, the adoption of Lunda narratives throughout this western and southern region was an outcome of the Rund capital becoming the hub of the 18th and 19th century slave trade. This interpretation of the later spread of Lunda titles is not universal; Ndaywel è Nziem, for example, remains committed to the historicity of Kinguri and the earlier dating system.29 For present-day Congolese Lunda intellectuals, of even more importance than the dating and chronology, the history confirms the Kinguri titleholder as the perpetual uncle to the mwant yav (see perpetual kinship discussed under Ethnographic Fieldwork and Historical Linguistics), and thus the affiliation of Lunda-related groups in the DRC and Angola.30 Similarly, the Luvale of Zambia have imbricated themselves in the oral tradition, in part to claim autonomy from their powerful Lozi neighbors, and, as Robert Papstein puts it, to “re-establish contact with Inkalanyi [the proto-Rund state] and its successor state, Luunda [Rund].”31 Appreciating such critical revisions of the historicity of the oral tradition holds implications for the extent of the Lunda commonwealth and how closely tied it was to the slave trade emanating from Angola. In particular, an earlier date suggests Rund expansion independent of the slave trade; a post-1700 expansion suggests that the Rund kingdom and Lunda system was tied to the Atlantic slave trade.

Questions regarding Kinguri and Kinyama’s exodus extend to the blood kin connection between the Lunda hunter-hero Chibind Yirung and his alleged grandfather, the Luba founder, Mbidi Kiluwe. Perhaps this connection was a later addition to the narrative that emerged as trade routes began to cross from the continent in the 18th century. (Luba traditions, for example, do not include the departure of Chibind Yirung.) If these were fictive rather than real kin, the assumed linear spread of Luba political institutions into Lunda polities can be questioned. Such reinterpretations also conform to the historiography of other related polities. Nineteenth-century rivalry between the Kanyok chiefs and the Luba mulopwe inspired aspects of the Kanyok tradition of genesis of Citend, a humiliated Luba princess who is said to have founded the Kanyok kingdom.32 The slave and ivory trade to the east, which Anne Wilson contended was key to the Luba-Katanga expansion, could equally have imbricated Luba oral traditions, similarly challenging the historicity of their characters along with chronology and causation.33 Where documentary verification exists, as in the nearby Kongo kingdom, John Thornton suspects modern oral traditions reflected the trade routes of the 19th century rather than Kongo in the 16th and 17th centuries.34 Linguistic evidence, further explored below, also suggests the exchange of Luba and Lunda political titles rather than a spread of Luba institutions. Jeffrey Hoover, who conducted lengthy fieldwork that informed his analysis of the Rund tradition, concludes that although the Luij/Chibind Yirung story was not simple fabrication, it is not certain that Luij and Chibind Yirung were historical individuals. Even as the story probably represents a moment when certain Luba individuals influenced the early Rund polity, Hoover finds that the Rund polity owed less to Luba government than Vansina and others previously thought.35

The slow-moving debate over central African oral traditions suggests that scholars need to consider the context of their performance and transcription to better appreciate what traditions of origin tell us. In addition to reflecting the prestige of trading hubs and of religious centers, a reconstructive approach to oral traditions, such as that applied by Pier Larson to Malagasy traditions, demonstrates how oral traditions helped to reconstitute kin networks in the face and in the aftermath of the slave trade by building of coherent narratives and celebrating founder kings.36 Many of the key informants in the reconstruction of Luba history were missionary converts, former slaves and refugees from slave-trading wars, who gathered around missions and told their stories to missionaries such as Harold Womersley and William Burton in the early 20th century.37 In the case of the Kuba, stories of Imbangala and Portuguese slave traders, probably disseminated through market trade in the 19th century, came to be told in the Kuba oral tradition as a motive for Kuba migration and consolidation.38 For the Rund, the first detailed written record of the oral tradition by Major Henrique de Carvalho published in 1890 followed the conquest of the Rund by Chokwe slave and ivory traders. Carvalho’s informants, were, as Victor Turner describes, “émigré aristocrats, who dwelt wistfully on their former grandeur.”39 For the Luba, Lunda, and others, literate colonial or protocolonial agents gave the opportunity to former (or changing) elites to reconstruct and romanticize history.

As conquest–mercantile polities emerged during the 19th century, legitimacy became a key concern in oral traditions. Narratives of legitimacy hinged on relationships between representatives of the Luba and Lunda systems on the one hand and ritual owners of the land on the other. In the Lunda case, a kilolo (cilolo) representative ruled a tribute-bearing village alongside a pre-Lunda owner of the land, the mwin ngand, who retained ritual control over the land.40 In the Luba case, the oral tradition has the religious and ritual authority of the owners of the land incorporated into the mulopwe titleholder. Still the bambudye titleholders, as those who owned the narratives and rituals of the Luba, remained distinct from the political authority of a mulopwe and from a man of power without ritual investiture known as a mukalanga.41

Foreign political cultures linked to the extension of regional and international trade further challenged the relationships between political overlords and ritual owners of the land. The most striking case is that of Msiri, the head of a group of Nyamwezi traders, who migrated to southeastern DRC in search of ivory and copper. He established the Yeke polity, which was based out of his capital of Bunkeya, a convergent point for trading routes from the east and west coasts. Msiri managed to forge a number of local alliances, first with chief Katanga, along with trading ties to Zanzibari and Angolan Luso-Africans, in particular the Ovimbundu. His local rule, in particular over the Sanga, became represented in the Yeke oral tradition as a promise to protect them from the incursions of foreigners. There proved, however, a contradiction between Msiri’s Nyamwezi (Sumbwa) political culture that combined the role of ruler over people and the ritual owner of the land in the Nyamwezi paramount (the mwami) and the Luba/Lunda system that identified distinct political overlords from local ritual owners of the land.42 The Yeke oral tradition placed such emphasis on legitimacy and autochthoneity, perhaps precisely because the Yeke were challenged in this regard. They appropriated common stories. A striking example is the story of Chief Kapema’s sister who requested Msiri to kill her brother, Kapema, since he had killed her son. Msiri kills Kapema, and this episode acts as charter and justification for Msiri’s rule. The story is an exact replica of the eastern Lunda oral tradition of the Kazembe kingdom, except for a change in names and titles. Yet the interpretation and historical significance of the stories differ. In Msiri’s case, it forms the basis for his local legitimacy; in Kazembe’s case, it justifies Lunda aristocrats as agents of Kazembe’s rule.43

Within the oral traditions, praises celebrate—and occasionally critique—kings and kingship. Hyperbolic praises were composed and performed by those invested in the kingdom. How should the historian treat the praise for the mwant yav, recited by Rund notables to Carvalho, declaring his omnipotence even after he had been defeated by the Chokwe?

We are no more than thy humble slaves; thou art the lord of our bodies, our lives, of our wealth, of all that thou surveyest.

If we are great, thou art greater than we are; there is no one above thee

O creator who makest arms and legs, look on all and see no one

Thou, o lord of rivers, trees, stones, all land, all lives, verily all is thy possession, lord!44

Not all praises were so celebratory. Sometimes they reveal tragic weaknesses or flaws of character. With careful interpretation, historians glimpse the political morality that constrained and critiqued kingship.45 For the most part, however, critique has given way to praise. This is especially apparent in the rituals that buttress kingship and that survived the 19th century. The elaborate prostrations when greeting a king, the royal dances of conquest, and death and burial rituals are typically viewed as proof of the power of kings.46 Consider, however, that these rituals came to be encouraged through the violence of the 19th century, and ossified into performative tradition during the colonial period.

Before proceeding to other forms of evidence, it is worth summarizing what is revealed by these transcribed oral traditions. Historians gain an understanding of the intellectual history of central African kingship, the culture upon which the centralization of power rested. Oral traditions contain ideas about competition over succession, especially between matrilines and patrilines, a logical element in any hereditary political system such as kingship. Symbolic qualities revolve around death, fecundity, and fertility, key concerns of political power that draw on intellectual themes of great antiquity. These symbolic qualities are unlikely to have emerged only with the rise of centralized hereditary titles but were certainly a key aspect of them. Oral traditions might also reveal perceived affiliations between people with the increased reach and influence of 19th-century trade. More difficult to establish, especially from traditions of origin, are absolute chronologies; nor do these court-oriented or even lineage-focused narratives indicate the reach of polities or the precise mechanisms of power and its enforcement. They may reveal more about the turmoil of the 19th century and efforts to construct a perceived legitimate form of centralized power than of older moments of expansion, contraction, or even the reasons behind these dynamics.

Art and Material Culture

The history of central African kingdoms reaches an international audience through their renowned art. Traditional art, like oral tradition, needs to be appreciated in terms of the place and period of creation. Two types of collections exist: (1) the material culture unearthed through archaeological digs, and (2) art collected by missionaries, colonial officials, ethnographers, travelers, and tourists since the late 19th century.

Some of the earliest central African art to attain widespread recognition in the early 20th century was that of Kuba statues of kings, the ndop, collected by American missionary William Sheppard and ethnographers Leo Frobenius and Emile Torday. These statues were initially thought to have been carved to celebrate Kuba kings during their reign or perhaps to commemorate them soon after. Subsequently stylistic analyses by art historians suggest that most of the classic ndop statues were all carved contemporaneously, probably in the late 18th century. The consensus is that the Kuba (that is the royal Bushoong clan) king who reigned in the late 18th century had the statues of his predecessors commissioned to enhance the prestige of the royal court. The creation of these statues may have been one indication of the transition to kingship from smaller-scale chiefship. Still, dating the ndop statues remains elusive, despite the apparent certainty of museum labels.47

Figure 3. Ndop portrait of King Mishe miShyaang maMbul (reigned early 18th century). Dated by Ronsenwald and Cornet to c.1760–1780, wood and camwood powder, dimensions 49.5 x 19.4 x 21.9 cm. Brooklyn Museum, 61.33. Creative Commons License.

The ndop statues, along with Torday’s celebration of the Kuba, led the Belgian colonial government to view the Kuba as a special civilization, a kingdom worth protecting through indirect rule.48 Even as the Kuba lay on the outskirts of the Luba-Lunda system, with closer cultural and linguistic affiliations to the north, through the celebration of their art (including Elisofon’s famous photograph at the beginning of this article), they have provided a model, even an archetype of central African kingship.

Archaeological findings through the 1960s provided insight into the historical antiquity of royal artefacts. The most influential site for the Luba-Lunda system involves excavations of graves at the Upemba depression, near the center of the Luba affiliates. The sites indicate social hierarchies and degrees of political centralization over approximately 1,500 years. Pierre de Maret, the archaeologist responsible for many of the sites and their analysis, finds six general periods: the Kamilambian (7th century ce), with iron agricultural implements, characterized by villages with low population densities; the ancient Kisalian to the 9th century, with a greater concentration of people and ceremonial goods (including decorated axes that have some resemblance to modern Luba axes), showing signs of centralized religious and political power; the classic Kisalian, the 10th–12th centuries with an increase in population, ceremonial goods combined with trade goods (copper, ivory, and cowries), indicating Indian Ocean trade along with a concentration of grave goods suggesting a rich minority class; the Kabambian A in the 13th–15th centuries with fewer iron goods and symbols of power but monetary units, such as copper crosses and cowries, perhaps evidence of the collapse of state structures in the face of trade; and finally, the Kabambian B, in the 16th–18th centuries, characterized by an increase in luxury trade—in particular copper smelting. While more recent graves were not intentionally excavated, a few graves identified as Luba were dated to around 100 years prior to the 1970s.49

The question is whether the earlier artifacts of wealth and prestige indicate kingship and, specifically, were precedents for Luba kingship. Oral traditions suggest the centralization of Luba polities was linked to the control of salt and iron, as well as the production of cloth from raffia palm.50 However, the depth of Iron Age artifacts along with the abundance of luxury goods in the Kisalian and Kabambian period suggests that regional trade based on salt and iron preceded Luba consolidation. Aspects of the material culture of the region are over 1,000 years old, in particular iron weaponry, ceremonial axes, and hoes.51 Such iron weapons and tools thus long preceded the rise of specifically Luba polities. Evidence of religious rituals and the concentration of settlements also show continuity: filing of teeth, and demographic density are all similar. Burial rites, on the other hand, change with the Luba in around the 18th century. There are several possible conclusions. Ndaywel è Nziem emphasizes continuity: according to this view, Luba origins were far earlier than indicated by the oral tradition and kinglists.52 Much depends on what we identify as origins. Political transformations over the last 1,500 years—including centralization and decentralization—long preceded the Luba polities of the 19th century. De Maret associates the Luba oral tradition, in particular the defeat of Nkongolo by Kalala Ilunga, with either the emergence of the Kisalian (9th–12th century), or the transition from Kisalian to Kabambian (13th–15th century) or perhaps the arrival of eastern traders in the 19th century (all of these periods could have inspired distinct episodes in the narrative).53 There was continuity and change over this period. De Maret, however, indicates that in the 18th and 19th centuries there was a decisive change, leading to the end of the Kabambian B period, and the rise of a specifically Luba state. Luba began as an ideological construct, drawing on earlier Kisalian and Kabambian traditions, and then, perhaps with the importation of a foreign political structure, “expanded during the nineteenth century to become of the largest states in Central Africa “This development,” de Maret emphasizes, “proceeded in step with that of long-distance trade with the Arabs.”54

Similar conclusions could be developed for other instances of central African kingship to the west. For example, the Feti and related stone ruins in the plateau of southwestern Angola, which were never properly excavated, indicate an important precedent for Ovimbundu kingship: even as Ovimbundu political transformation proceeded through long-distance trade with the Atlantic world.55

Prestigious (or royal) Luba art collected in the 19th and 20th centuries does not correspond closely to that found at archeological sites. Perhaps predominantly wood objects have not survived, or perhaps they were not the type of item included in grave goods. Nonetheless, the lack of continuity in what has become defined as Luba royal art does not allow a firm establishment of its historical depth. Like the Kuba royal statues discussed above, the dating of stylistic elements of 19th-century kingdoms remains vague. Still, this internationally celebrated art, including stools, staffs, bow stands, sculpted figurines, and mnemonic devices such as the lukasa boards (see Figure 2), portray aspects of the cosmology of kingship.

Few other genres of African art reveal this much focus on the female body: coiffures, navels, genitalia, and scarifications (in particular on the staffs of ruling elites) (Figure 4), caryatid stools (Figure 5), and sculpted figurines used for divination (Figure 6), made of wood, bone, ivory, and even scarified directly onto the skin.

Figure 4. Wooden Luba mulopwe Staff (kibango), 19th century. Dimensions 149.9 x 9.5 x 9.8 cm. Brooklyn Museum, 22.1132. Creative Commons License.

Figurines, stools, and staffs in the shape of women were in different ways receptacles for the spirits of male kings: Stools evoked the very structure of a kingdom; staffs positioned the female founders closest to the life-giving sky; figurines mobilized invisible powers of fecundity; and scarified patterns held esoteric meanings and contributed to sexual pleasure. As Mary Nooter Roberts puts it, for the Luba this suggests that “women are receptacles of spiritual energy and beholders of political secrets . . . Indeed, their very bodies present a ‘discourse on power’ to those permitted to see them.”56 This art, combined with interpretations of the oral traditions, suggest that the supernatural qualities of Luba kingship were concerned with powers over fertility and overcoming sterility in people and land.57

Figure 5. Luba Female Caryatid Stool. © The Metropolitan Museum of Art.

Image source: Art Resource, NY, ART324241.

Figure 6. Luba Divination Bowl. Early 20th century. Dimensions 56.2 cm. Royal Museum for Central Africa, Tervuren, Belgium.

The Luba functioned firstly as a fertility cult, mediated by the bambudye society that spread across villages and came to appoint certain male leaders, prophetic mulopwe, responsible for fertility and fecundity. The pre-19th century Luba system would thus be better thought of as a religious movement concerned with fertility, which at certain times and places came to uphold the rule of hereditary rulers legitimized by this spiritual system. Art and oral tradition reveal more about this religious system and this fertility cult than the construction of a secular state. In this sense, the emphasis on Luba kingship might be misplaced; so, is it more accurate to think of the early Luba system as the extension of a priestly class, with a hierarchy that extends from the bambudye society to the mulopwe? As anthropologist Pierre Petit contends, there were religious foundations to Luba kingship; in fact, it is likely that Luba was first a religious movement and then transformed into a secular-style system of hereditary titles with the commercial and security imperatives that began in the 18th century.58

In most Luba- and Lunda-related polities, female political titles existed, some of which were the equivalent of “queen mothers,” such as the mwadi for the Luba and the nswana mulopwe (or simply Luij) for the Rund. These were female titles; in the Luba system women occupied them. In the Lunda perpetual kinship system, however, men could hold a female title or vice versa—although, after Luij, there is no record of a woman occupying the supreme title of mwant yav and possessing the insignia of rule.

Surprisingly given the prominence of Luij in the Lunda oral tradition, there is no similar celebration of the female body in Lunda art. Much of the Lunda art represented in international collections can be connected to Chokwe caravan merchants, who by the late 19th century conquered and dominated the nuclear Rund court. Their iconic representations of Lunda kingship include the famous masculine representation of the hero of the oral tradition, Chibind Yirung (Chibinda Ilunga, as in Figure 7). Provenance is difficult to establish, and there is much discussion over what actually constitutes Chokwe art. It is unclear whether the most representative figurines were carved by the Chokwe or emerged from their trade with Lunda artist and clients. The most authoritative art historian, Marie-Louise Bastin, identifies such art as part of an “expansive” phase of Chokwe polities from the mid-19th century, as opposed to earlier court art.59

Figure 7. Chibinda Ilunda Statue of Lunda Founding Hero. Attributed to the Chokwe, who dominated the Lunda court by the late 19th century. Dimensions 40.6 x 15.2 x 15.2 cm. Kimbell Art Museum, Fort Worth, Texas.

The Chokwe seemed concerned to inscribe themselves into the prestige of Rund historical heroes. The Chokwe were intrepid mobile traders, adept at drawing on multiple styles to affirm the powers of their leaders. In addition to Rund influences, they adapted widespread stool construction to imitate Portuguese chairs, thus creating “thrones” for their kings.60 Such thrones became popular across the region, probably spreading along trade routes. Among the Pende for example, they were items of status rather than functioning seats.61 In one remarkable example of Chokwe appropriations, a carving at the back of the throne vividly represents the conflict between Luij and her brothers, Kinguri and Kinyama.62 The throne pictured in Figure 8 is adorned by brass tacks imported from Europe that had begun to replace copper from Katanga. With its back framing a Chihongo ancestral mask—the male equivalent of the Mwana Pwo mask (Figure 9), used in male initiation rites and in succession rituals—the throne joins local forms of legitimacy such as control over initiation to the power and swank acquired through their access to international trade items such as brass.

Figure 8. Chokwe Throne, 19th Century. Copper alloy, animal hide, wood, 26 3/4 x 12 x 15 1/2 in. (67.9 x 30.5 x 39.4 cm). Brooklyn Museum, Museum Expedition 1922, Robert B. Woodward Memorial Fund. Creative Commons License.

As with the Chibinda Ilunga (Chibind Yirung) figure, the antiquity of such representations is unknown. The art objects were generally collected in the early 20th century; it is unlikely that they predate Chokwe-Portuguese interactions. The Portuguese-influenced thrones suggest, like the revisionist history of the oral narrative, that the art of Lunda-Chokwe kingship was entwined with the 18th and 19th century slave trades.

The masking culture, typical of the Chokwe and nearby (including Lunda) affiliates and the Pende, represent an older albeit ever-changing form of religious and political aesthetics. Used by men, the masks represented ancestors (makishi), sometimes women, as in the Chokwe mask of the Mwana Pwo ancestor in Figure 9. Masquerades generally mobilized horizontal forms of political affiliation during performances, often linked to the initiation of boys (mukanda), in which sodalities of age sets were established. However, they could also celebrate the lineage of rulers, as illustrated by the Chihongo mask on the throne pictured in Figure 8, or be performed in secular ceremonies for entertainment and education (as seems to have been the case for the Mwana Pwo masks). By the 20th century these masks had spread across Chokwe, Lunda, and Luvale polities.63 Mwana Pwo masks emphasized the importance of maternal figures, although based on the existing literature, they seem less focused on fertility than the Luba figurines. Lunda–Chokwe masks were autonomous and older forms of political affiliation, even if kings, too, attached their authority to these sodalities.

Figure 9. Chokwe Mask (Mwana Pwo), Early 20th Century. Dimensions, 30.5 x 21.5 x 22.7 cm. Brooklyn Museum, Gift of Corice and Armand P. Arman. Creative Commons License.

To conclude, aspects of material culture, including iron and copper items, predated Luba and Lunda kingship. Archaeology suggests that moments of political centralization—kingdoms perhaps—were evident prior to the Luba and Lunda systems. The Luba and Lunda consolidation, this evidence suggests, was likely tied to the extension of trade routes, confirming aspects of the revisionist history of the oral traditions already discussed. The proliferation of Luba and Chokwe art, to name the most prominent found in international collections, occurred during what Marie-Louise Bastin terms the “expansive” moment of the 19th century. Evidence suggests the emergence of distinctive arts, in particular the celebration of the female figure for the Luba and the hunter-hero for the Chokwe, which then spread as status and religious objects across the region. Luba political clients sought insignia of power, such as bow stands, staffs, and stools. Chairlike thrones, ornamented with characters from the popular Lunda oral tradition, supported big men, prophets, and warlords associated with the slave trade to Angola. Such men celebrated the hunter-hero Chibinda Ilunga, seeking figurines of him as power objects. Masks in this region indicate older but dynamic forms of political affiliation, of sodalities, even if they, too (like the Luba religion), become entwined with the power plays of the 18th and 19th centuries.

Ethnographic Fieldwork and Historical Linguistics

Historical evidence garnered by contemporary fieldwork can be “upstreamed” into historical time, even as historians have to take care not to introduce anachronistic elements. Chronologies from such evidence tend to be at best relative unless related to other forms of evidence. Two types of fieldwork-based evidence describe aspects of kingship: ethnographic investigations into political culture on the one hand and the collection of words of modern languages, and then their reconstruction through historical linguistics, on the other. Both contribute to knowledge of concepts of leadership, succession, and political organization.

Ethnographic reports based on fieldwork began in the early 20th century. In the Belgian Congo, colonial officials conducted ethnographic inquiries and sometimes referred them back to Belgium.64 Occasionally, they published findings in the Bulletin des Juridictions Indigènes et du Droit Coutumier Congolais (BJDCC).65 Missionaries produced ethnographic reports and occasionally monographs on customs and traditions, many of which became sources for historians.66 Reports from the colonial administration detailed forms of succession, inheritance, and the division of labor—key concerns for a colonial administration. Colonial ethnographers began to identify a “belt” of people who traced their descent through their matrilineal kin (i.e., through association with their mother’s siblings). Political titles, in particular among the most powerful polities, were distributed across these matrilineages, often by means of patrilineal succession. Conflicts between matrilineages often resulted. These conflicts proved a headache for the colonial administration, a source of fascination for anthropologists—and in the broader and longer view, a political feature of hereditary titleholders.

During the colonial era in British-ruled Northern Rhodesia (Zambia), the Rhodes-Livingstone Institute scholars advanced the professional anthropological understanding of concepts of kingship, succession, inheritance, and traditional government. Audrey Richards and Godfrey and Monica Wilson applied the concept of divine kingship to the Bemba and Nyakusa respectively; Victor Turner described the symbolic elements of Lunda-Ndembu cosmology, and Max Gluckman wrote about law among the Lozi, to name only a few of the most influential.67 Ian Cunnison’s study of perpetual kinship and positional succession among the Eastern Lunda, the kingdom of Kazembe, made inroads into understanding the problem of titular and royal succession and proved influential for later historiography. Building on the work of Audrey Richards among the Bemba, Cunnison identified a system of succession of name and titles (positional succession) along with the inheritance of kin relationships of those who occupied these titles, which he termed “perpetual kinship.” Mwata Kazembe, the Eastern Lunda paramount, for example, inherited the totality of his predecessors’ identities and kin relationships.68

The combination of perpetual kinship and positional succession was not found across the region. Different forms of succession and clientage prevailed among Luba polities. Reefe’s ethnographic investigations, building on work conducted by missionaries Pierre Colle and William Burton, identified the extension of Luba clientage through confirmation rites that invested titleholders with spiritual power. The Luba king, the mulopwe, distributed insignia—bracelets, canes, waistbands, and war implements—to signify attachment to the Luba system. In turn, mediums and diviners sent sacred white earth (lupembe for the Luba; such white clay enabled spiritual power throughout the region) to the kings, and this was in turn given to clients. It is, however, difficult to determine whether this system helped to strengthen the Luba center or distributed power from that center.69 Insignia could be inherited, but their inheritance did not in itself legitimize new rulers. Aspects of positional succession existed for Luba-related societies, but Reefe and others found no system of perpetual kinship as elaborate as that of Lunda-related polities.

Based on his ethnographic work, Jeffrey Hoover detailed the distribution of Rund titles. These include among others the Rund title of mwant yav at the helm, his district administrators, the kilolo (or chilol) (Kanyok) found in the capital (Musumb) and across the territory; the female co-sovereign nswan murund (literally meaning “mother of the Rund,” also known informally as Luij); and the nswan mulapw (the successor to the mulopwe [Luba]); mwad[i] (first wife); namwana (mother of chief) along with several military titleholders, including the kazembe (expedition leader); the kalala (leader in vanguard); and the mazembe (leader in rearguard), as well as governmental functionaries such as the kakwata (enforcer).70 Rund titles were related to each other in a system of perpetual kinship often defined in the oral tradition. Thus nswan murund/Luij remained the mother of the king; and kinguri remained the maternal uncle of the king.

The historical question is how this system of perpetual kinship among the Rund emerged and evolved. Historical linguistics has helped to provide some answers. A caveat is needed, however. The proliferation of exchanges of political terms across south-central Africa, along with the uneven transcription of the many diverse languages, has slowed down reconstruction based on historical linguistics. To be sure, Vansina fruitfully applied the words-and-things approach to the decentralized political tradition of the equatorial forest and the political history of west-central Africa. To the south, Kathryn de Luna considered terms of leadership for the decentralized Bantu Botatwe speakers. The history of south-central African kingdoms offers distinct challenges, however. Since the introduction and proliferation of certain royal titles along with concepts of governance dates from the 18th century, the archaic semantic fields of such words may not be as revealing for the particular polities concerned. In many cases, the antiquity of concepts of governance and titles is unknown. In situations where linguists have offered more specific focus on royal titles for centralized kingdoms, as Boeston and colleagues do for the ngangula title of the Kongo kingdom, documentary evidence anchors such words within the kingdoms in the 17th century.71 No such evidence exists for the Lunda and Luba polities. Still, careful reconstruction has provided some relative chronology and opened up investigation into the early history and precursors of the Lunda and Luba systems.

Hoover was one of the first scholars to supplement his ethnographic observations with linguistic reconstruction, yet his findings regarding the origins and evolution of perpetual kinship were not conclusive. He hypothesized that perpetual kinship provided the ideological glue that cemented relationships between individual Rund titleholders by countering the divisive tendencies of their increasingly separate lineages. Perpetual kinship may have emerged, Hoover thought, with the declining role of genealogically based kinship: the role of blood descent groups giving way to the perpetual kinship of political titles.72 Vansina further developed this analysis to reconstruct forms of collective governance in “early” Kasai: that is, before the rise of the Rund (which Vansina dates to c. 1600). Based on common terms in Kasai’s proto-language, Vansina argues that government in early Kasai revolved around bilateral houses connected by men’s associations consisting of ranked membership with masks appropriate to each rank. This system of houses and sodalities informed the later political structure of greater Kasai, most notably the Rund and Kuba, but in different ways. According to Vansina’s linguistic analysis, the Rund transformed rule by association into rule by a king. The Rund kingdom became a “single house writ very large.”73 Associational sodalities and houses developed into a polity in which titles became related through perpetual kinship. Vansina thereby confirms Hoover’s hypothesis that the political problem the Lunda resolved was that of succession of houses: Instead of the collapse of a house at the death of its leader, by making the successor to the leadership into their predecessors in all regards (i.e., through perpetual kinship), the structure of the kin government was preserved. A bilateral house, probably drawing on the governmental resources of associational sodalities, developed into a kingdom based on perpetual kinship.74 An enterprising historical linguist may continue these investigations; up to the present, Hoover and Vansina’s preliminary work remain the only linguistic reconstruction of the specifically Luba and Lunda politico-religious systems.

The timing of the development of these political institutions still remains vague. Furthermore, without historical anchors, it is difficult to reflect on the possible causes for such centralizing processes of political power. Were they internal and rooted in a fascination with political power, as Vansina implies? Or can they be related to material aspects? In addition to the regional and international slave trades, the control over salt, fish, iron, and copper may have, at different times, contributed to political centralization. From the 16th century, global changes impacted on these processes. Ahead of the direct influence of the Atlantic slave trade, Brazilian cassava (manioc) spread, replacing older crops such as millet. In northwestern Zambezi watersheds, among the Lunda-related polities on the southwestern borders of Rund influence, Achim von Oppen observed how cassava provided the opportunity for greater differentiation between men and women, and between ranks of men, allowing for processes of political consolidation. Men could become involved in export production (essentially the slave trade) and also wanted female slaves to work in the fields and reproduce families limited by matrilinearity (with matrilineal affiliations the children of men belonged to their wive’s brother; men could, however, claim the children of their slaves).75 The association of cassava with a praise for one of the early Luba rulers, Ndayi Mwine Nkombe, also suggests the influence of cassava on the growth of the Luba-Lunda system.76 Maize (corn) and cassava also spread to the Kuba around the time that the Bushoong clan consolidated their rule.77 Fieldwork in economic anthropology thus suggests that the Atlantic world, through the slave trade and the arrival of American crops, may have catalyzed political developments. Linguistic investigations into the words adopted for foreign foods and crops could offer further insight.78

In conclusion, ethnographic evidence, along with the study of words and political titles, reveals how the Lunda diffused power through the inheritance of titles and perpetual kinship between such titles. The Luba on the other hand inherited objects of power; titleholders were vested with sacred powers that contributed to fertility and fecundity.79 Conflict over politico-religious insignia was not directed against a central Luba state; rather, competition seemed to be over becoming part of it.80 Luba was a religious and spiritual resource, mobilized by diverse titleholders. In terms of actual political consolidation, the Luba developed lineages combined with associations, in particular the bambudye; whereas the Lunda, out of associational sodalities and bilateral houses, developed the fiction of lineage (i.e., perpetual kinship) as a basis for governance and for far-flung alliances.

Why did Luba and Lunda clients want to remain part of the system? Objects, insignia, titles, and masks, rather than affirmation of an exclusive center, signified power. Conflict over such paraphernalia and prestigious titles could be centripetal and centrifugal—even at the same time. But what happened when external forces such as the slave and ivory trade attracted clients away from the center and toward other centers of wealth and power? Documentary sources provide the most direct evidence of the processes that influenced much of the region from the late 18th century.

Documentary Evidence

Reports by traders, explorers, and missionaries for the interior of the continent, west of the Kwango River and East of Lake Tanganyika, begin to emerge at the end of the 18th century, a period that historians have characterized as marked by the “decline” of the kingdoms. For many polities, however, this was also a period of expansion. It may also be described as a period of reconstructive effort to project the cohesion of the past in the face of the violence of the present. If kingdoms were principally acts of the imagination, when exactly did they exert a monopoly over force? Or are they better thought of as fertility cults and extended fictive families that only took the form of protective kingdoms in the 19th century, even as they recollected a glorious and violent past in the search for legitimacy? Documentary evidence helps to explore some of these possibilities.

The first documentary evidence of note is De Lacerda’s expedition of 1798 to the Eastern Lunda of Mwata Kazembe, the Kazembe kingdom of Luapula. De Lacerda recorded the rise of the Kazembe kingdom some sixty years prior, dating the spread of Lunda titleholders to this area to c. 1740. A fascination for Portuguese and Luso-African traders in the early 19th century, the Luso-African merchants, Baptista in 1806–1809, and Gamitto in 1831–1832 also describe the Kazembe kingdom.81 De Lacerda, Baptista, and Gamitto describe the Kazembe kingdom probably at its most powerful: its capital was surrounded by extensive defensive ditches, protected by several thousand armed men, and inhabited by at least ten thousand.82 After 1860s, records from Zanzibari Hamed ben Mohammed el-Murjebi (Tippu Tip), David Livingstone, Victor Giraud, and early colonial accounts indicate the relative decline in the reach of the kingdom.83 The Kazembe kingdom is a rare case of being able to crosscheck the oral record with early-19th-century Portuguese and late 19th-century European documentary sources.

For other Lunda-related polities, the documentary record is sparse and generally restricted to the latter half of the 19th century. Luso-African merchants and the British missionary explorer David Livingstone first accounted for other western Lunda–related polities, with some detail on the Rund.84 German colonist Paul Pogge’s descriptions of the customs and oral traditions of the Rund were followed in even greater detail by the influential descriptions of Portuguese explorer and advocate Major Henrique Augusto de Carvalho.85

There are even fewer records of the Luba court. Gamitto in 1832 and the Livingstone in 1868 mention the Luba without much detailed description. The first substantial source is Verney Cameron, who spent five months at the court of an heir to the Luba throne in 1874–1875.86 The memoirs of the Tippu Tip, who raided and traded with the Luba around the 1870s, complement the few European accounts.87 By the 1890s, these were followed by colonial officials and army officers linked to the Congo Free State (CFS), along with Catholic and Protestant missionaries. Each descriptive account has distinctive qualities, from the self-interested trader to careful ethnographer. Some accounts are unreliable: They are rife with familiar clichés of African savagery, even cannibalism; and the depictions of violence could be self-serving justifications for European conquest.88 Yet, as Richard Reid has indicated for east Africa, these accounts provide invaluable details about the military revolution of the 19th century—and, for central Africa, the dynamics of trade as well.89

When considered as a whole, an image of concentrated violence emerges, not ubiquitous across the region but around caravan routes, expanding new polities and collapsing old ones. De Carvalho provides copious material on the Rund in the aftermath of conquest by Chokwe.90 For the Luba court of Kasongo, Cameron provides a rich description from his visit in 1874–1875, when he witnessed a brutal ruler mutilating his subjects for minor punishments and pleasure and relying on his trading alliances with the Zanzibari Jumah Americani and Bihé-based Luso-African merchants.91 Livingstone and Tippu Tip provide further evidence of the conquest of Luba areas by Zanzibari traders, Msiri’s Yeke, and Luso-African Ovimbundu traders, dissipating the power of their Luba religious imagination and forcing the emergence of protective and violent polities.92 Missionaries and early colonial records depict the Bemba warrior state led by the Crocodile Clan.93 Along with early European mercenary recruits of the CFS, missionaries also record the crisis of Yeke ruler Msiri’s state. CFS officers assassinated Msiri; his precarious polity, already weakened by a Sanga rebellion, collapsed.94 Similarly, the archives of European officers contain extensive records of the Songye/Tetela hero (but never a king), Ngongo Luteta, executed by his European comrades after using them to rebel against his previous Zanzibari masters.95

The need for protection from trade and its violent consequences ratcheted up the military element of politics. This process was reached at different times for different polities. By the mid-19th century, both the Lunda and Luba polities sought to monopolize trade by limiting the activities of caravan routes beyond their realms and followers.96 They all engaged in international trade to access new and better guns and powder, even if the demand for guns and gun technology did not, as Giacomo Macola points out in his scholarship based largely on the documentary record, determine the political history of the region.97

In general, the violence and dynamism recorded by 19th-century documentary sources hold implications for the interpretation of art and oral traditions. It confirms that these traditions and artifacts were produced in a political economy of raiding, trading, and enslavement. The violence recounted in the oral traditions and enacted during their performances—as in the renowned dances of conquest—depict the need for warriors and for violent leaders.98 Art, particularly Luba art, presents an idealized world, indicating a desire for harmony. Such representations were a product of aspirations for peace and fecundity in the face of the violence indicated in the documentary record.

In addition to shedding light on each other, the documentary and oral records are not discrete. Explorers and ethnographers wrote down the earliest versions of the oral traditions, sometimes, as in the case of De Carvalho for the Rund, in considerable detail. Missionaries and colonial officials would also later record these narratives. Sometimes such missionaries pioneered efforts to edit and publish local histories, as in Harold Womersley’s collection of Luba oral traditions. Alternatively, they facilitated the publication of vernacular authored “ethnohistories,” as in the Catholic White Father Edouard Labrecque’s role in the publication of Mwata Kazembe’s Ifikolwe fyandi na Bantu Bandi (My ancestors and my people), an editorialized version of the oral tradition.99 Both of these cases informed school textbook accounts of the history of the kingdoms. Scholars during colonial and postcolonial times further facilitated such histories, as Auguste Verbeken did for Antoine Munongo and the history of the Yeke during colonial times and Robert Papstein for the Luvale in postcolonial Zambia.100 Alternatively, histories could emerge out of subnational ethno-projects, as in Jason Sendwe’s and Bonaventure Makonga’s articles on the history and culture of the Luba, Joseph Okito’s record of the traditions of Ngongo Luteta, and the son of Msiri, Mukanda Bantu’s writings on the Yeke.101

Beyond Kingdoms

By the time Vansina published Kingdoms of the Savanna, the diverse forms of political authority of south-central Africa—rendered in oral tradition, art, ethnography, colonial and missionary records, and ethnohistories—coalesced around the concept of kingship. Not only was kingship politically convenient to newly independent African countries claiming their sovereignty, but it also drew on these diverse evidentiary resources. Kingship was not a false construction. The focus on it, however, only allowed for a blurred outline of the broader political and religious culture of the region. This final section considers the implication of conceiving of political arrangements in terms of kingship, along with alternative ways of rendering these forms of government and politico-religious power. Diverse hierarchies and affiliations were described as “kingdoms” due to the presence of hereditary titleholders who at some point claimed control over land, people, or the ability to command—and sometimes through a sense of their spiritual agency. Usually titles were some sort of combination of all of these traits. Hereditary titleholders led diverse forms of political and religious associations: fertility cults, extended fictive family, and mercenary groups. A single titleholder could be all of these at once, or have their most distinctive features change throughout their lifetimes.

Depending on how we characterize it, the Lunda and Luba politico-religious systems are of some antiquity, at least as old as the 13th century. However, as kingdoms identifiable by terms similar to Luba and Lunda with kinglists that refer to actual historical personalities, they seem to have emerged between the 17th and 18th centuries. Regarding the nuclear Lunda, the Rund, even if the lineage of the mwant yav may be older, its characteristics as a kingdom had not coalesced prior to 1700. The documentary record dates the expansion of the Rund kingdom itself to around 1700, although the extent to which this was a colonizing thrust or an adoption by peripheral groups of Rund political titles (or an exchange of political titles between periphery and center) is unclear. Nonetheless, through the 18th century a number of polities east of the Kwango and west of the Lualaba and Luapula rivers began to identify themselves as part of the Lunda extended family: the Lunda commonwealth. It seems likely that the rise of these affiliations were linked to extending patterns of trade (probably the Atlantic slave trade) or the spread of new crops such as Brazilian manioc (cassava), or a combination of these factors.

The history of Luba polities is even more vague, especially if the relationship of Chibind Yirung, who is said to have initiated the Rund dynasty, is not considered the grandson of Luba hero Mbidi Kiluwe. The documentary record establishes an extensive Luba polity in the early 19th century that in the Katanga region took the form of a kingdom ruled by a mulopwe titleholder. Even given the Luba tendency to legitimize themselves through a celebration of antiquity, and potentially add ancestral deities to the kinglist, the origins of the mulopwe lineage date to c. 1700.102 But, like the Rund, we have little idea of the number of political subjects, the strength of their armed forces, their wealth, or territory at that time. How many people actually identified themselves by a term similar to “Luba,” and what did such an identity mean?

By tracing the exchange of political terminology, the consolidation of certain political and religious structures around the 17th century seems apparent. For Lunda-related polities these structures included the inheritance of political titles related to one another through a system of perpetual kinship. The Luba, on the other hand, were associations of men who celebrated women, or, put differently, claimed to hold secrets and power over fecundity and fertility. They may have originated as hunting associations, experts in the secrets of nature, or associations of healers who promoted the fertility of people, or both.103 These associations manipulated symbols of power, including imaginatively embellished titles, which were significant sources of legitimacy.104 These religious ideas were of some antiquity; prophets who then became kings (or warlords) along with their followers molded and disseminated them through the 18th and 19th centuries.

Luba and Lunda forms of association were not mutually exclusive. They were not tribes, ethnolinguistic communities, or exclusive political identities. A titleholder who formed part of the Lunda fictive family could adopt the Luba religious system. They could be a Lunda titleholder and a Luba prophet—or a Luba prophet could turn into a Lunda titleholder. Indeed, given the extensive borrowings back and forth from Luba and Lunda, especially pertaining to concepts relating to political titles and religion, the intermingling of Luba and Lunda language and oral traditions (rather than the distinctiveness of Luba and Lunda as tribes or even distinct polities) is the most likely scenario. For a few polities, such as the Kanyok or even the Kazembe kingdom, the mixing of Lunda and Luba titles is obvious. Hoover makes clear, however, that Luba and Lunda titles were exchanged and intermixed even in the heart of the Rund kingdom.105 The early documentary evidence also indicates the intermixture of Luba and Lunda with Kasongo, the Luba mulopwe (by then more of a warlord than a prophet or sacred king) informing Cameron in 1875 that the Rund mwant yav was also Luba. This confirmed the contemporaneous importance of the relationship between Mbidi Kiluwe and Chibind Yirung in the Rund oral tradition.106

Centers and peripheries were difficult to delineate in the 19th century and probably earlier. Msiri had created the center of his Yeke polity out of the periphery of the Kazembe kingdom.107 Those on the apparent periphery of the Rund, such as the Imbangala titleholder of Kinguri, insinuated themselves into the center of the Lunda story of origin; Kinguri became Luij’s brother and the mwant yav’s maternal uncle. By the 19th century, with the centers of military power and wealth shifting toward the west and east coasts, the polities of the interior (previously thought of as centers) were dependent on the periphery. Even in the oral traditions, power was seen to come from the east or west. There may have been a cosmological element to this conception: the rising and setting of a sun that contributes to drought or fecundity, as de Heusch claims. However, the fact is the polities became beholden to the power of the caravan traders from the east and west.108 It is fairly clear that by the late 19th century the Luba of Kasongo relied on mercenaries tied to the caravan trade rather than internal forms of leigtimation.109

In the tumult of 19th century (in some places earlier, as for, say, the Kazembe kingdom), the need for protection from the slave and ivory trade—and, in turn, the need to participate in that trade—converted older political and religious affiliations into political hierarchies that claimed a monopoly over violence. Precarious identities solidified into political subjectivities. A good example is the Bemba polity, which by the late 19th century had consolidated around the most violent of the Crocodile Clan leaders. However, the oral tradition and insignia of rule at the heart of its cosmology was still oriented around the Luba fertility cult.110 Stories of origins, symbolic and idealized renderings told in the 19th century, were narratives of violence that held the promise of peace and fertility. They reveal the appropriation of older political and religious ideas and institutions by warlords. Msiri’s Yeke state relied on powerbrokering among Luba and Lunda along with alliances with international traders; his oral tradition emphasized the legitimacy of his violent defense of local chiefs, even as such local chiefdoms remember Msiri’s violence differently.111

Warlords mobilized objects and symbols of power in support of hierarchical forms of political organization. Should such warlords be considered as kings? The most recent historian of Msiri, Hugues Legros, eschews the term “warlord” as the product of Western prejudice and describes the Yeke polity as a kingdom (“royaume”). However, the office of Msiri was never succeeded prior to colonialism and was not entrenched in deep-seated tradition. For Legros, the vital argument in favor of considering Msiri a divine king (roi sacré) is that he underwent rituals of investiture for a Nyamwezi mwami.112 On the other hand, warlords such as Ngongo Luteta never went through such rituals. Tetela and Songye, however, would in later years claim that Luteta was a founding king. Historians should exercise caution. The recognition of some rituals and forms of memorialization, which are highly political and linked to processes of legitimization, is arbitrary. There are no objective and universal diagnostic rituals that indicate kingship.113 Perhaps instead it is better to identify such rulers by their practices. In this sense, Msiri and Ngongo Luteta were both warlords. Still, through colonial and postcolonial ethnic mobilizations, these founding heroes have become memorialized as kings.

The relationship between violence and state building was not always evident. Decentralized agents of violence also proliferated: the Waungwana, Ngongo Luteta’s warriors, the Ruga Ruga, the Balungu, the warlord Lusinga, and the Chikundu are some examples. If violent movements did not consolidate into colonial tribal sovereignties they disappeared from the documentary record, as was the case with the Balungu.114 Others, such as Ngongo Luteta’s followers and the Waungwana and Ruga Ruga, lost their specific identities as they were incorporated into colonial armed forces.115 The warlord Lusinga’s head found its way to a Belgian museum; his soldiers and subjects at first helped to form a Roman Catholic kingdom and then Tabwa ethnicity.116 The Chikunda slave soldiers became a marginal ethnicity.117 Non-violent political structures that did not align with the colonial language of kingdoms and tribes became known as “secret societies,” either ceasing to exist or practiced outside the purview of the missionaries and colonial state.118 Local and decentralized lineages did not conform easily to the colonial politics of tribalism and are underrepresented in the historiography of kingdoms.119

By the late 19th century, powerful leaders and their followers exchanged titles, rituals, art, words, and narratives. Their transformation into tribes through the colonial period facilitated their transition into supposedly archaic kingdoms. Diverse ethnonyms became standardized into tribal names and then kingdoms: Rua, Arua, Luwa into Luba; Awewa, Wewa, Awemba into Bemba; Ruund, Ruwund, and Aluunda into Lunda; Quiboco, Quiboque, Ciboku became Cokwe, Chokwe, or Tchokwe. Minor linguistic differences allowed for an assertion of difference: Some participants in the Luba religious cult, refugees led by a Luba warrior, who settled by the Lulua River (from Rua?) became, for example, Luluwa or Lulua, eventually constituting their own kingdom.

These late-19th-century reconstructions were not historical inventions but rather adaptations that conformed to political exigencies by concentrating power in titles. Older political cultures, perpetual kinships, and religious movements came to be imagined as kingship. At a certain point, as enough followers—and, later, powerful outside patrons (African, Zanzibari, and Euro-African traders, missionaries, explorers, colonists, anthropologists, and historians)—came to believe in this vision of kingship, it became a historical reality. The key to understanding central African kingship is not found in its origins, as much of the previous historiography emphasizes, but in its supposed decline. For its moment of eclipse became that of its celebration; the artifacts of this celebration form its evidentiary base.

In the violent reordering of central Africa through the colonial and postcolonial periods, kingdoms remained useful but for different reasons and with different outcomes. The Lulua mentioned above is an instructive example. With the cementing of tribal identities in the colonial period, the Lulua became enemies of the Luba-Kasai.120 Changes in postcolonial provincial boundaries and shifting political alliances has, however, joined the Luba-Kasai and the Lulua in the Kamwina Nsapu rebellion of 2016–2017. This follows conflict over the state appointment of the Luba-Kasai titleholder of the same name and then the assassination of the popular heir. The Kamwina Nsapu militia has reconstituted the Lulua–Luba into a new political unit, largely in opposition to the postcolonial state but also battling Chokwe-Pende rivals.121 Ancient history will resurface as new alliances and friendships are revealed in a dynamic, relevant, and contested historical narrative. The historiography of the central African kingdoms is up for grabs.

Primary Sources

The following is a summary of those sources mentioned and discussed in each section of the article.

    Oral Traditions

    Since the late 19th century, oral traditions of Luba, Lunda, and related polities have been recorded on numerous occasions. For the Luba-Katanga, the most thorough version is Womersley’s, as discussed by Reefe.122 For the Rund, Henrique Augusto Dias de Carvalho and Léon Duysters is the best resource, as discussed by Hoover.123

    A first generation of literate elites recorded traditions in colonial-era publications. These include Jason Sendwe and Bonaventure Makongo for the Luba, Joseph Okito for the Tetela, Mukanda Bantu for the Yeke, and Jacques Chileya Chiwale and Mwata Kazembe XIV for the Kazembe kingdom.124

    Art and Material Culture

    Luba and Lunda art is now scattered across numerous international public and private collections. The single most important and comprehensive collection of royal art is held by the Royal Museum of Central Africa, Tervuren, Belgium.125 The most thorough archaeological investigation into the precursors of the Luba system is the investigation of grave goods in the Upemba depression, with findings reported by Pierre de Maret.126

    Ethnographic Fieldwork and Linguistics

    Ethnographic investigation into the history and structure of the Luba and Lunda polities flourished during the colonial period. Missionary ethnographic investigations especially valuable for Luba history include Pierre Colle and William F. Burton.127 Belgian colonial officials also undertook research. Although archives in the Democratic Republic of Congo are no longer functioning, many ethnographic reports can be found in the ethnographic section of the RMCA at Tervuren. Local-level investigations in response to ethnographic questionnaires can be found in the Belgian African Archives in Brussels. Some colonial reports were published in the Bulletin des Juridictions Indigènes et du Droit Coutumier Congolais. Colonial officials who published valuable work include E. D’Orjo de Marchovelette, Edmond Verhulpen, and Léon Duysters. In British-ruled Zambia (Northern Rhodesia), colonial officials undertook occasional historical investigations relating to chieftaincy and succession, with their reports found in the National Archives of Zambia. The colonial state also sponsored the Rhodes-Livingstone Institute anthropologists, whose extensive work was published by the Rhodes-Livingstone Institute in monograph form or in the Rhodes-Livingstone Journal. Several of these anthropologists have deposited fieldnotes in archives or have allowed researchers to access them if privately held.128 Numerous dictionaries provide data on words and help in linguistic reconstructions.129 For Rund titles and concepts of governance consult Hoover’s Appendix 2.130

    Documentary Sources

    These include sources written prior to colonialism by traders, explorers, and soldiers involved in conquest. Many have been published, including De Lacerda, Baptista, Gamitto, Hamed ben Mohammed el-Murjebi (Tippu Tip), Livingstone, Giraud, Wissman, Pogge, Cameron, and de Carvalho. Published memoirs and accounts of officials linked to King Leopold’s Congo Free State, including, among others, Stanley, Dhanis, Delcommune, Le Marinel, Verdick, Brasseur, and Briart, should be supplemented with their unpublished correspondence and papers held at the RMCA Historical Section.131

Further Reading

  • Bastin, Marie-Louise. La sculpture tshokwe. Meudon, France: A. and F. Chaffin, 1982.
  • de Heusch, Luc. The Drunken King, or The Origin of the State. Bloomington: Indiana University Press, 1982.
  • de Maret, Pierre. “The Power of Symbols and the Symbols of Power Through Time: Probing the Luba Past.” In Beyond Chiefdoms: Pathways to Political Complexity in Africa. Edited by Susan Keech McIntosh. Cambridge, UK: Cambridge University Press, 1999.
  • Hoover, Jeffrey. The Seduction of Ruwej: Reconstructing Rund History (The Nuclear Lunda; Zaire; Angola; Zambia). PhD diss., Yale University, 1978.
  • Legros, Hugues. Chasseurs d’Ivoire: Une histoire du royaume yeke du Shaba (Zaïre). Brussels: Editions de l’Université de Bruxelles, 1996.
  • Macola, Giacomo. The Gun in Central Africa: A History of Technology and Politics. Athens: Ohio University Press, 2016.
  • Ndaywel é Nziem, Isidore. Histoire générale du Congo: De l’héritage ancient à la République Démocratique. Paris and Brussels: De Boeck & Larcier, 1998.
  • Reefe, Thomas Q. “The Societies of the Eastern Savanna.” In History of Central Africa. Edited by D. Birmingham and P. Martin. Vol. 1. New York: Longman, 1983.
  • Reefe, Thomas Q. Rainbow and Kings: A History of the Luba Empire to 1891. Berkeley: University of California Press, 1981.
  • Roberts, M. Nooter, and Allen F. Roberts, Memory: Luba Art and the Making of History. Munich: Prestel-Verlag, 1998.
  • Vansina, Jan. Kingdoms of the Savanna. Madison: University of Wisconsin Press, 1966.
  • Vellut, Jean-Luc. “Notes sure le Lunda et la frontiere luso-africaine (1700–1900).” Etudes d’histoire africaine 3 (1972): 65–69.
  • Wilson, Anne. “Long Distance Trade and the Luba Lomami Empire.” Journal of African History 13, no. 4 (1972): 575–589.
  • Yoder, John C. The Kanyok of Zaire: An Institutional and Ideological History to 1895. Cambridge, UK: Cambridge University Press, 1992.


  • 1. From Christraud M. Geary, “Africans and Photography,” In and Out of Focus: Images from Central Africa, 1885–1960, ed. Migs Grove (London: Philip Wilson, 2003), 101.

  • 2. For the most recent account of decentralized forms of political leadership on the southern fringe of this region, see Katherine M. de Luna, Collecting Food, Cultivating People: Subsistence and Society in Central Africa (New Haven, CT, and London: Yale University Press, 2016), and for the Congo River basin to the north, Jan Vansina, Paths in the Rainforest: Toward a Political Tradition in Equatorial Africa (Madison: University of Wisconsin, 1990).

  • 3. A vague boundary, as in Jan Vansina’s delineation of the “Age of Chiefs” from the “Age of Kings” for the Kuba in The Children of Woot: A History of the Kuba Poeples (Madison: University of Wisconsin Press), 104–171.

  • 4. Thomas Q. Reefe, “The Societies of the Eastern Savanna,” in History of Central Africa, eds. D. Birmingham and P. Martin (Longman, 1983), 160–204; and Isidore Ndaywel é Nziem, Histoire générale du Congo: De l’héritage ancient à la République Démocratique (Paris and Brussels: De Boeck & Larcier, 1998), 129–161.

  • 5. Jan Vansina, “Forward,” in Mary Nooter Roberts and Allen F. Roberts, Memory: Luba Art and the Making of History (Munich: Prestel-Verlag), 13. For an equivalent idea for the Lunda, such as “une affaire mentale, politico-culturelle,” see Matadiwamba Kamba Mutu, Espace lunda et identities en Afrique centrale: lieux de mémoire (Paris: L’Harmattan, 2009), 15–16.

  • 6. David M. Gordon, “(Dis)Embodying Sovereignty: Divine Kingship in Central African Historiography,” Journal of African History 57, no. 1 (2016): 47–68.

  • 7. See, for example, Claude Meillassoux’s account of the rise of kingship and slave hierarchies in The Anthropology of Slavery: The Womb of Iron and Gold, trans. Alide Dasnois (Chicago: University of Chicago Press, 1991).

  • 8. For reading lukasa boards, see M. Nooter Roberts and Allen F Roberts, Memory: Luba Art and the Making of History (Munich: Prestel-Verlag, 1998), 140–142.

  • 9. Thomas Q. Reefe, Rainbow and Kings: A History of the Luba Empire to 1891 (Berkeley: University of California Press, 1981), 24–31.

  • 10. For episodes in different recordings, see Reefe, Rainbow and Kings, 32–37. It is unclear the extent to which Womersley’s account from the 1950s, intended as a popular text, may have influenced the rendering of the oral tradition by Reefe’s informants. The “inclusivity” might be a product of this feedback. See Harold Womersley, Legends and History of the BaLuba (Los Angeles: Crossroads, 1984).

  • 11. Jan Vansina, Kingdoms of the Savanna (Madison: University of Wisconsin, 1966), 71–72.

  • 12. See correspondence in the Vansina Collection, Herskovits Library, with publishers on dated nature of Kingdoms.

  • 13. Luc de Heusch, The Drunken King, or The Origin of the State, trans. Roy Willis (Bloomington: Indiana University Press, 1982), 42.

  • 14. Reefe, Rainbow and Kings, 44.

  • 15. Reefe, Rainbow and Kings, 80–92. For oral tradition inscribes in landscape also see David M Gordon, “History on the Luapula Retold: Landscape, Memory, and Identity in the Kazembe Kingdom.” Journal of African History 47, no. 1 (2006): 21–42.

  • 16. Jeffrey Hoover, in his still unpublished PhD thesis, provides the most thorough critical historical analysis of the myth, in “The Seduction of Ruwej: Reconstructing Rund History (The Nuclear Lunda; Zaire; Angola; Zambia)” (PhD diss., Yale University, 1978), 24–28. Fewer versions have been recorded compared to the Luba myth, perhaps a result of different missionary contexts; however, an earlier version was comprehensively recorded in Henrique Augusto Dias de Carvalho, Ethnographia e Historia Tradicional dos povos da Lunda (Lisbon: Impresa Nacional, 1890), 50–112. Portions of Carvalho’s account translated and annotated by Victor Turner, “A Lunda Love Story and its Consequences: Selected Texts from Traditions Collected at the Court of Mwatiamvwa in 1887,” Rhodes-Livingston Journal, 1–26. Other versions include Léon Duysters, “Histoire des Aluunda,” Problèmes d’Afrique Centrale 40, no. 12 (1958): 75–98.

  • 17. Vansina, Kingdoms, 97.

  • 18. Ndaywel è Nziem, Histoire, 138.

  • 19. See the critique of De Heusch’s symbolic interpretation of the Lunda oral tradition, Jeff Hoover, “Mythe et Remous Historique: A Lunda Response to de Heusch,” in History in Africa 5 (1978): 63–80.

  • 20. Manuela Palmeirim, Of Alien Kings and Perpetual Kin: Contradiction and Ambiguity in Ruwund (Lunda) Symbolic Thought (Wantage: Sean Kingston, 2006), 20–56.

  • 21. De Carvalho first noted that Kinguri provided the possibility for dating the Lunda kingdom to the 16th century; Turner agreed with de Carvalho’s dating; see Turner, “A Lunda Love Story,” 15–16. Hoover first claimed Kinguri was a later addition to the tradition; see Hoover, “Seduction of Ruweij,” 211–243. Through independent analysis, John Thornton agreed with Hoover but emphasized documentation in John Thornton, “The Chronology and Causes of Lunda Expansion to the West,” Zambia Journal of History 1 (1981): 1–13.

  • 22. The revisions suggested below invalidate much of Vansina’s Kingdoms, 78–97.

  • 23. Joseph Miller, “The Imbangala and the Chronology of Early Central African History,” Journal of African History 13, no. 4 (1972): 549–574. Joseph Miller develops a more ambitious reinterpretation of the Lunda oral tradition but is less firm on dating in Kings and Kinsmen: Early Mbundu States in Angola (Oxford: Oxford University Press, 1976), 112–150.

  • 24. Jean-Luc Vellut, “Notes sure le Lunda et la frontiere luso-africaine (1700-1900),” Etudes d’histoire Africaine 3 (1972): 65–69.

  • 25. Jan Vansina, “It Never Happened: Kinguri’s Exodus and Its Consequences,” History in Africa 25 (1998): 387–403.

  • 26. While written before the revision of Kinguri’s exodus, Robert Edmond Schechter’s analysis of oral traditions among the Zambezi and other Lunda polities confirms the autonomous development of Lunda polities, rather than a colonizing thrust of the central Lunda; see Robert Edmond Schechter, “History and Historiography on a Frontier of Lunda Expansion: The Origins and Early Development of the Kanongesha” (PhD diss., University of Wisconsin-Madison, 1976).

  • 27. In the most thorough study of Luvale traditions, Robert Papstein accepts the historicity of the departure of Chinyama from Rund, basing his dating on number of capitals occupied by each holder of the Chinyamma title: to the end of the 15th century. However, even if as with Kinguri the title is of some antiquity, it could have been a 19th-century addition to the Rund epic. See Robert Papstein, “The Upper Zambezi: A History of the Luvale People, 1000–1900” (PhD diss., University of California-Los Angeles, 1978), 116–123.

  • 28. Vansina, “It Never Happened,” 401.

  • 29. Ndaywel è Nziem Histoire, 153. Part of the problem, it seems, is that much of this revisionist scholarship is in English and has not fully penetrated French historical scholarship.

  • 30. Matadiwamba Kamba Mutu, Espace lunda, 36–37.

  • 31. Papstein, “The Upper Zambezi,” 123.

  • 32. John C. Yoder, The Kanyok of Zaire: An Institutional and Ideological History to 1895 (Cambridge, UK: Cambridge University Press, 1992), 29–80. For further study of Kanyok kingship, based in structural analysis but with detailed reference to documentary sources, see Rik Ceyssens, Le roi Kanyok au milieu de quatre coins (Fribourg, Switzerland: Èditions Universitaires, 2003).

  • 33. Anne Wilson, “Long Distance Trade and the Luba Lomami Empire,” Journal of African History 13, no. 4 (1972): 575–589.

  • 34. John Thornton, “Modern Oral Traditions and the Historic Kingdom of the Kongo,” in The Power of Doubt: Essays in Honor of David Henige, ed. Paul Landau (Madison: University of Wisconsin-Madison, 2011), 195–207.

  • 35. Hoover, “Seduction,” 183–205. Hoover still finds an earlier “phase” of Luba influence on Rund, presumably prior to Rund influence on the Luba. Still, Hoover claims that the Rund system was “fundamentally their own creation, not a conquest state which derives its inspiration, forms, and legitimacy from the Luba or anyone else,” with only a few Luba terms, representing a “relatively peaceful synthesis through intermingling populations and through communication over time.” (205). For the historicity of the Luij/Chibinda Ilunga story, see 227–243.

  • 36. Pier M. Larson, “Reconsidering Trauma, Identity, and the African Diaspora: Enslavement and Historical Memory in Nineteenth-Century Highland Madagascar,” William and Mary Quarterly 56, no. 2 (April 1999): 335–362.

  • 37. David Maxwell, “The Creation of Lubaland: Missionary Science and Christian Literacy in the Making of Luba Katanga in Belgian Congo.” Journal of Eastern African Studies 10, no. 3 (2016): 367–392. As mentioned, Harold Womersley provided the most thorough rendering of the Luba myth, as well as writing episodes to be used for Congolese school textbooks; see Harold Womersley, Legends and History of the BaLuba (Los Angeles: Crossroads, 1984). William Burton’s work, especially Luba Magic and Religion and Magic in Custom and Belief (Tervuren, Belgium: Musée Royal de l’Afrique Centrale, 1961), as well as the artifacts he collected, remain key resources.

  • 38. Vansina, Children of Woot, 36–37.

  • 39. Henrique de Carvalho, Ethnographia, 58–112; Victor Turner translates and annotates this part of Carvalho’s work in Victor Turner, “A Lunda Love Story and its Consequences: Selected Texts from Traditions Collected at the Court of Mwatiamvwa in 1887,” Rhodes-Livingston Journal, 19 (1955): 1–26, 1.

  • 40. Hoover, “Seduction,” 102–103. The relationship of control over land versus people depended on the nature of the land. For scholarship on the resource-rich lagoons and fertile lands of the Luapula and heightened role of owners of land and lagoons, see Gordon, Nachituti’s Gift, chapter 1. For work on Lunda overlords and owners of land, see Matadiwamba Kamba Mutu, Espace lunda, 107–108.

  • 41. Hoover, “Seduction,” 101–113; Reefe, Rainbow, 41–48.

  • 42. For migration, settlement, and legitimizing oral traditions, see Hugues Legros, Chasseurs d’Ivoire: Une historire du royaume yeke du Shaba (Zaïre) (Brussels: Editions de l’Université de Bruxelles, 1996), 27–59; for more on the contradictions between Sumbwa and Luba/Lunda political culture regarding political sovereignty and ritual ownership, see 46–47.

  • 43. For the narrative and interpretation of Msiri and Kapema, see Legros, Chasseurs, 14–15, 42–43. But contra Legros, the eastern Lunda oral tradition does not have Kazembe killing Nkuba but the Lunda aristocrat, Kalandala; see Gordon, Nachituti’s Gift, chapter 1.

  • 44. Carvalho, Ethnographia, 74; Turner, “A Lunda Love Story,” 13.

  • 45. See, for example, the royal praises of the eastern Lunda, in Jacques Chileya Chiwale, Royal Praises of the Lunda Kazembe of Northern Rhodesia: Their Meaning and Historical Background (Lusaka, Zambia: Rhodes-Livingstone Institute, 1962).

  • 46. David M. Gordon, “The Cultural Politics of a Traditional Ceremony: Mutomboko and the Performance of History on the Luapula,” Comparative Studies in Society and History 46, no. 1 (2004): 63–83; and see David M. Gordon, “History on the Luapula Retold: Landscape, Memory, and Identity in the Kazembe Kingdom,” Journal of African History 47, no. 1 (2006): 21–42.

  • 47. Vansina, Children of Woot, 211–215; Monni Adams, “18th Century Kuba King Figures,” African Arts 21, no. 3 (May 1988): 32–38, 88.

  • 48. For indirect rule and the Kuba, see Jan Vansina, Being Colonized: The Kuba Experience in Rural Congo, 1880–1960 (Madison: University of Wisconsin Press, 2010), 178–209.

  • 49. The comprehensive account in Pierre de Maret, Fouilles archéologiques dans la vallée du Haut-Lualaba, Zaire- Sanga and Katongo, 1974, 2 vols. (Tervuren, Belgium: Musée Royal de l’Afrique Centrale, 1982) and Fouilles archéologiques dans la vallée du Haut-Lualaba, Zaire—Kamilamba, Kikulu, et Malemba-Nkulu, 1975, 2 vols. (Tervuren, Belgium: Musée Royal de l’Afrique Centrale). A synthesis essay is found in Volume 3, chapters 1–3, 177–231. Much of that synthesis is summarized in English in Pierre de Maret, “The Power of Symbols and the Symbols of Power Through Time: Probing the Luba Past,” in Beyond Chiefdoms: Pathways to Political Complexity in Africa, ed. Susan Keech McIntosh (Cambridge, UK: Cambridge University Press, 1999).

  • 50. Reefe, Rainbow, 98.

  • 51. De Maret, Fouilles, 227; “Power of Symbols,” 155.

  • 52. Ndaywel è Nziem, Histoire Générale, 130, 138.

  • 53. De Maret, Fouilles, 229–231.

  • 54. De Maret, “Power of Symbols,” 159; and De Maret, Fouilles, 227–228.

  • 55. Jan Vansina, How societies are Born: Governance in West Central Africa Before 1600 (Charlottesville: University of Virginia Press, 2004), 162, 170–174.

  • 56. M. Nooter Roberts and Allen F Roberts, Memory: Luba Art and the Making of History (Munich: Prestel-Verlag, 1998), 112. Also see Mary Nooter Roberts, “The King is a Woman: Shaping Power in Luba Royal Arts,” African Arts 46, no. 3 (2013), 68–81.

  • 57. De Heusch, Drunken King, 34–75; and see Bemba, 229–245 and Gordon, Invisible Agents, chapter 1.

  • 58. Pierre Petit, “‘Les charmes du roi sont les esprits des morts’: Les fondemonts religieux de la royauté sacrée chez les Luba du Zaire,” Africa 66, no. 3 (1996): 349–366.

  • 59. Marie-Louise Bastin, Art Décoratif Tshokwe (Lisbon: Companhia de Diamantes de Angola, 1961). Also see updated discussion and summary by Baris Wastiau, “Style and Ethnicity: Reflections on Methods for the Study of Arts in the Zambezi and Kasai Headwaters,” in A Anthropologia dos Tshokwe e povos aparantados: Colóquio em homenagem a Marie-Loise Bastin, ed. Manuela Palmeirim (Porto, Portugal: University of Porto, 2003), 49–78.

  • 60. Marie-Louise Bastin, La sculpture tshokwe (Meudon, France: A. & F. Chaffin, 1982), 251; and Reinhild Kauenhoven-Janzen, “Chokwe Thrones,” African Arts 14, no. 3 (May 1981): 69–74, 92.

  • 61. Z. S. Strother, Visions of Africa: Pende (Milan: 5 Continents Editions, 2008), 41.

  • 62. Reinhild Kauenhoven-Janzen, “Chokwe Thrones,” 72–73.

  • 63. Manuel Jordan, ed., CHOKWE! Art and Initiation Among Chokwe and Related Peoples (Munich: Prestel, 1998), and for typology of masks, see Manuel Jordan, “Zambian Makishi Masquerades and the Story of Categories,” in A Anthropologia dos Tshokwe e povos aparantados: Colóquio em homenagem a Marie-Louise Bastin, ed. Manuela Palmeirim (Porto, Portugal: University of Porto, 2003), 79–97.

  • 64. For example, for the Luba, see the colonial official Edmond Verhulpen, Baluba et Balubaïsés du Katanga (Anvers: Editions de L’Avenir Belge, 1936); and for the Lunda, Duysters, “Histoire des Aluunda.”

  • 65. For example, the work of E. D’Orjo de Marchovelette, “Notes sur les funérailles des chefs Ilunga Kabale et Kabongo Kumwimba: Historique de la chefferie Kabongo,” BJDCC 18, no. 12 (1950): 350–368; D’Orjo de Marchovelette, “Historique Kongolo: Histoire de la chefferie Kongolo” 19, no. 1 (1951): 1–13; and D’Orjo de Marchovelette “Quelques considerations sur les Bambudie du territoire de Kabongo,” BJDCC 8, no. 10 (1940): 275–289.

  • 66. Missionaries were especially influential in Luba history, including Pierre Colle, Les Baluba 2 (Brussels: Albert Dewit, 1913); and Burton, Luba Religion and Magic; Womersley, Legends and History of the Baluba.

  • 67. Godfrey Wilson, The Constitution of Ngonde (Livingstone, Northern Rhodesia: Rhoses-Livingstone Institute, 1939); Monica Wilson, Divine Kings and the “Breath of Men” (Cambridge, UK: Cambridge University Press, 1959); Audrey Richards, “Keeping the King Divine,” Proceedings of the Royal Anthropological Institute of Great Britain and Ireland (1968), 23–35; Max Gluckman, The Judicial Process among the Barotse of Northern Rhodesia (Glencoe, IL: Free Press, 1955); Victor Turner, The Drums of Affliction: A Study of Religious Process among the Ndembu of Zambia (Oxford: Clarendon, 1968); and Victor Turner, The Forest of Symbols: Aspects of Ndembu Ritual (Ithaca, NY: Cornell University Press, 1967).

  • 68. Ian Cunnison, “Perpetual Kinship: A Political Institution of the Luapula Peoples,” Rhodes-Livingstone Journal 20 (1955): 28–48.

  • 69. Reefe implies both, in Rainbow, 79–92.

  • 70. Hoover, “Seduction,” 101–113; Note extensive reconstruction of meaning and origin of titles in Appendix 2, 527–577.

  • 71. Koen Bostoen, Odjas Ndonda Tshiyayi, and Gilles‐Maurice de Schryver, “On the Origins of the Royal Kongo Title Ngangula,” Africana Linguistica 19 (2013): 53–83.

  • 72. Hoover, “Seduction,” 117–121.

  • 73. Jan Vansina, “Government in Kasai before the Lunda,” International Journal of African Historical Studies 31, no. 1 (1998): 1–22, 20.

  • 74. Vansina, “Government in Kasai.” Vansina’s historical argument regarding the origins of perpetual kinship do not sit easily with the symbolic argument of its functioning, as outlined by Palmeirim, Alien Kings and Perpetual Kin, 57–69. Elaborating on the historical linguistics of political terminology, Jan Vansina further considers the remote history of the origins of the early Kasai system in How Societies Are Born.

  • 75. Achim von Oppen, Terms of Trade and Terms of Trust: The History and Contexts of Pre-colonial Market Production Around the Upper Zambezi and Kasai (Münster, Germany: Lit Verlag, 1993), 258–293. Jan Vansina traces cassava to caravan routes, “Histoire du Manioc en Afrique Centrale avant 1850,” Paideuma 43 (1997): 255–279.

  • 76. Reefe, Rainbow, 59, 62.

  • 77. Vansina, Children of Woot, 175–177.

  • 78. For Kiswahili and Portuguese loanwords that spread along trade routes, including food and crops, see David M. Gordon, “Localizing the Global: The Wanderwörter of Nineteenth-Central South Central Africa,” in Tracing Language Movement in Africa, eds. Ericka Albaugh and Kathryn de Luna (Oxford: Oxford University Press, 2018), 259–276.

  • 79. But this division is not precise; for example in the stories of Lunda expansion, the Kazembes or migrating royals carry insignias of rule that with them, such as drums and marimba, according to the oral tradition in, H. Van Roy, “L’origine des Balunda du Kwango,” Aequatoria 24 (1961): 136–141.

  • 80. Reefe, Rainbow, 92.

  • 81. The influence of these documentary sources, and their role in reconstructing history of the kingdom, Giacoma Macola, Kingdom of Kazembe: History and Politics in North-Eastern Zambia and Katanga to 1950 (Münster, Germany: Lit Verlag, 2002), 2–8. The most accessible version of De Lacerda’s voyage can be found in Lands of Cazembe, ed. R. F. Burton (London: John Murray, 1875), 33–164. Voyage of the Luso-African merchant the pombeiros by Pedro João Baptista is most accessible in translation by B. A. Beadle and R. Burtons, trans., Lands of Cazembe (London: Royal Geographic Society, 1873). For the pombeiros, also see A. Verbeken and M. Walraet’s edited volume La Première Traversée du Katanga en 1806: Voyage des Pombeiros d’angola aux Rios de Sena (Brussels: Institut royal colonial belge, 1953). An accessible English translation is A. C. P. Gamitto (translated by I. Cunnison), King Kazembe, 2 vols. (Lisbon: Junta da Investigaçoes do Ultramar, 1960).

  • 82. Macola, Kingdom of Kazembe, 115–128.

  • 83. Macola, Kingdom of Kazembe, 115–128; and Victor Giraud, Les Lacs de l”Afrique Equatoriale: Voyage d’exploration exécuté de 1883 a 1885 (Paris: Librairie Hachette, 1890).

  • 84. For the Rund, see Joaquim Rodrigues Graça, “Expedição ao Muatayamvua: Diaro de Joaquim Rodrigues Graça,” Boletim da Sociedade de Geographia de Lisboa 9.8–9 (1890); and David Livingstone, Livingstone’s African Journal, 1853–1856, ed. Ian Schapera (Berkeley: University of California Press, 1963).

  • 85. Paul Pogge, Im Reches des Muato Jamwo (Berlin: Verlag von Dietrich Reimer, 1880); and Henrique Augusto Dias de Carvalho, Ethnographica e história tradicional dos povos da Lunda (Lisbon: Imprenza Nacional, 1890).

  • 86. Verney Cameron, Across Africa, 2 vols. (New York: Harper and Brothers, 1877).

  • 87. Tippu Tip’s autobiography published as Hamed bin Muhammed, Maisha ya Hamed in Muhammed el Murjebi yaani Tippu Tip Kwa maneno yake. Supplement to the East African Swahili Committee Journal 28, no. 2 (July 1958) and 29, no. 1 (January 1959, with helpful annotations, François Bontinck, trans. and ann., L’autobiographie de Hamed ben Mohammed el-Murjebi Tippo Tip (ca. 1840–1905) (Brussels: Académie royale des Sciences d’Outre-Mer, 1974).

  • 88. As Johannes Fabian describes, in Out of our Minds; Reason and Madness in the Exploration of Central Africa (Berkeley: University of California Press, 2000).

  • 89. Richard Reid, “Violence and its Sources: European Witnesses to the Military Revolution in Nineteenth-Century Eastern Africa,” in The Power of Doubt: Essays in Honor of David Henige, ed. Paul Landau (Madison: University of Wisconsin-Madison/Parallel Press, 2011), 41–59.

  • 90. Combined with other work, see Carvalho, Ethnographia.

  • 91. Verney Lovett Cameron, Across Africa (New York: Harper and Brothers, 1877), 298–348.

  • 92. Hamed bin Muhammed, Maisha ya Hamed, 95. See David Livingstone’s 1871 field diary for surprising versions of Zanzibari influence in Maniema, especially as recently transcribed by the UCLA Library through a spectral imaging project, allowing for comparison with published versions.

  • 93. Roberts, History of the Bemba, 2–12.

  • 94. Legros, Chasseurs, 129–158. In addition to oral tradition, Legros’s version is based on records of missionary F. S. Arnot, Bihé and Garenganze: A Record of Four Years’ Work in Central Africa (London: James E. Hawkins, 1893[1969], originally published 1889) and Daniel Crawford Thinking Black (London: Morgan & Scott, 1912); and G. E. Tilsley ed. Diaries of Dan Crawford. See the published memoirs and diaries of CFS officers, Paul Le Marinel, Carnets de route dans l’État Independent du Congo de 1887 à 1910 (Brussels: Éditions Progress, 1991); Alexandre Delcommune, Vingt années de la vie africaine; Edgard Verdick, Les premiers jours au Katanga; and articles in the pro-colonial Le Mouvement Géographique, especially C. Stairs, “L’Expedition Stairs,” Le Mouvement Géographique 16 (1892), 70. Relevant archives from officers at RMCA include papers of Francis Dhanis, Clément Brasseur, Paul Briart, and Edgard Verdick.

  • 95. RMCS, Dhanis Papers. As described by David M. Gordon, “Interpreting Documentary Sources on the Early History of the Congo Free State: The Case of Ngongo Luteta’s Rise and Fall,” History in Africa: A Journal of Method 41 (2014): 5–33; and for details on Ngongo Luteta, see David M. Gordon, “Precursors to Red Rubber: Violence in the Congo Free State, 1885-1895,” Past and Present no. 236 (2017): 133–168.

  • 96. As reported by Livingstone, in Schapera, African Journals, 228; Cameron, Across Africa (New York: 1877), 157–158. For Kazembe, the Angolan pombeiros were detained in the kingdom from two to four years, as recorded in the diary of Pedro João Baptista, which was then reprinted in an annotated French version, La Première traversée du Katanga en 1806: Voyage des “pombeiros” d’Angola aux Rios de Sena, ed. and trans. A Verbeken and M. Walraet (Brussels: Institut royal du colonial belge, 1953).

  • 97. Giacomo Macola, The Gun in Central Africa: A History of Technology and Politics (Athens: Ohio University Press, 2016).

  • 98. As in Palmeirim’s account of the investiture of the mwant yav, in Alien Kings and Perpetual Kin, 145–160.

  • 99. Giacomo Macola, “Literate Ethnohistory in Colonial Zambia: The case of Ifikolwe Fyandi na Bantu Bandi,” History in Africa 27 (2001): 187–201.

  • 100. For Yeke, see A. Verbeken, Msiri, roi du Garenganze (Brussels: L. Cuypers, 1956). Relationship with Antoine Munongo revealed in Verbeken Papers, RMCA. For Luvale, Moses Kaputungu Sangambo, The History of the Luvale People and Their Chieftainship (Los Angeles: Africa Institute, 1979).

  • 101. For Jason Sendwe and Bonaventure Makongo, see David Maxwell, “The Creation of Lubaland: Missionary Science and Christian Literacy in the Making of Lubaland in Belgian Congo,” Journal of Eastern African Studies 10, no .3 (2016): 367–392. Joseph Okito recorded the oral traditions while acting as conseiller provincial of Kasai in 1957, published as, “Notes historiques,” Communauté, nos. 10–21 (September 1957–March 1958). For Mukanda Bantu, see Legros, Chasseurs, 19, 219, 222.

  • 102. Reefe, Rainbow, 49–63.

  • 103. The longue durée root for the hunter-leader among Botatwe people to the south, K. de Luna, “Hunting Reputations: Talent, Individuals, and Community in Precolonial South Central Africa,” Journal of African History 53, no. 3 (2012): 279–299, and Collecting Food, Cultivating People, 93–130.

  • 104. De Maret, “The Power of Symbols,” 161–162.

  • 105. For exchanges of loanwords between Luba and Lunda languages, see Hoover, “Seduction,” 183–204.

  • 106. Cameron, Across Africa, 326, 361

  • 107. On this, and the broader transformation of centers to peripheries and vice versa, see Legros, Chasseurs, 91.

  • 108. For spatial orientation described in Luba and Bemba oral tradition, see De Heusch, Drunken King, 39, 235.

  • 109. Cameron, Across Africa, 289–360; and Reefe, Rainbow, chapters 13–14.

  • 110. For Bemba as Kingdom, see Andrew Roberts, A History of the Bemba: Political Growth and Change in North-eastern Zambia before 1900 (London: Longman, 1973), chapters 5–6. For fertility cult and religion of the Bemba, see Gordon, Invisible Agents, chapter 1.

  • 111. Legros, Chasseurs, 39–59.

  • 112. Legros, Chasseurs, 62–75.

  • 113. Gordon, “(Dis)Embodying Sovereignty.”

  • 114. Rik Ceyssens, Balungu: Constructeurs et destructeurs de l’Etat en Afrique centrale (Paris: L’Harmattan, 1998).

  • 115. For more on the incorporation of Waungwana and Ruga Ruga into German East African Schutzruppe see Michelle Moyd, Violent Intermediaries: African Soldiers, Conquest, and Everyday Colonialism in German East Africa (Athens: Ohio University Press, 2014); for Ngongo Luteta’s followers into Force Publique, see Gordon, “Precursors to Red Rubber.”

  • 116. Allen F. Roberts, A Dance of Assassins: Performing Early Colonial Hegemony in the Congo (Bloomington: Indiana University Press, 2012).

  • 117. Allen Isaacman and Barbara Isaacman, Slavery and Beyond: The Making of Men and Chikunda Ethnic Identities in the Unstable World of South-Central Africa, 1750–1920 (Portsmouth, UK: Heinemann, 2004).

  • 118. For example, the Ubutwa society around Luapula, Mwelwa Musumbachime, “The Ubutwa Society in Eastern Shaba and Northeast Zambia to 1920,” International Journal of African Historical Studies 27, no. 1 (1994): 77–99. Another example is the bambudye society, which has experienced something of a renaissance through association with the mulopwe and a smoother entry into Christian missionary experiences.

  • 119. See, the exception, which is a complicated history of the chieftaincies of southern Katanga based on oral tradition: Léon Verbeek, Filiation et usurpation: Histoire socio-politique de la region entre Luapula et Copperbelt (Tevuren, Belgium: RMCA, 1987).

  • 120. Bogumil Jewsiewicki, “The Formation of a Political Culture of Ethnicity in the Belgian Congo, 1920-1959,” in Creation of Tribalism in Southern Africa, ed. Leroy Vaill (Berkeley: University of California Press, 1991), 324–349; for the Lunda case, see Edouard Bustin, Lunda Under Belgian Rule: The Politics of Ethnicity (Cambridge, MA: Harvard University Press, 1975).

  • 121. For current perception of the history of Kamwina Nsapu titleholder, see Le systeme Kamwina Nsapu.

  • 122. Harold Womersley, Legends and History of the BaLuba (Los Angeles: Crossroads, 1984).

  • 123. Henrique Augusto Dias de Carvalho, Ethnographia e Historia Tradicional dos povos da Lunda (Lisbon: Impresa Nacional, 1890), 50–112. Portions of Carvalho’s account translated and annotated by Victor Turner, “A Lunda Love Story” ; and Léon Duysters, “Histoire des Aluunda,” Problèmes d’Afrique Centrale 40, no. 12 (1958): 75–98.

  • 124. Jason Sendwe, “Traditions et coutumes ancestrales des Baluba Shankadji,” Bulletin du centre d’Etude des Problèmes Sociaux Indigènes 24 (1954): 87–120; Bonaventure Makonga, “La Comprehension des coutumes,” Bulletin des Juridictions Indigènes et du Droit Coutumier Congolais 23, no. 64 (1955): 149–153; Bonaventure Makonga, “Samba-a kya-buta,” Bulletin des Juridictions Indigènes et du Droit Coutumier Congolais 16 (1948), 321–345; Mukanda Bantu, “Mémoires de Mukanda Bantu”; Chiwale, Royal Praises of the Lunda Kazembe; and Mwata Kazembe XIV and E. Labrecque, Central African Historical Texts II: Historical Traditions of the Eastern Lunda, I. Cunnison, trans. (Lusaka, Zambia: Rhodes Livingstone Institute, 1961).

  • 125. Only a small portion of the collection is on display; however, the entire collection can be accessed online.

  • 126. Pierre de Maret, Fouilles archéologiques dans la vallée du Haut-Lualaba, Zaire- Sanga and Katongo, 1974, 2 vols. (Tervuren, Belgium: Musée Royal de l’Afrique Centrale, 1982) and Fouilles archéologiques dans la vallée du Haut-Lualaba, Zaire—Kamilamba, Kikulu, et Malemba-Nkulu, 1975, 2 vols. (Tervuren, Belgium: Musée Royal de l’Afrique Centrale).

  • 127. Pierre Colle, Les Baluba; and William F. Burton, Luba Religion and Magic.

  • 128. For Audrey Richards’s valuable fieldnotes on Bemba, see LSE Library. Another example is Victor Turner’s research notes for the Lunda-Ndembu, recently available through Alexander Street/ProQuest.

  • 129. These dictionaries include: M. K. Fisher, English-Lunda (Ndembu) Abridged Dictionary (Mutshatsha, Congo: M. K Fisher, 1970–1979); Malcolm Brook MacJannet, Chokwe-English, English-Chokwe Dictionary and Grammar Lessons (Vila Luso, Angola: Missão da Biulu, 1927, 1947); White Fathers, Bemba-English Dictionary (Ndola, Zambia: Mission, 1991); William M. Morrison, Dictionary of the Tshiluba Language (Luebo: Congo: J. L. Wilson, 1939); and Em Jenniges, Dictionnaire Français-Kiluba (Brussels: Spineaux, 1909), available online.

  • 130. Hoover, “Seduction,” 527–577.

  • 131. For an inventory of these private papers, see Patricia van Schuylenbergh, La Mémoire des Belges en Afrique Centrale: Inventaire des Archives Historiques Privées du Musée Royal de l’Afrique Centrale de 1858 à nos jours (Tervuren, Belgium: RMCA, 1997).