David L. Schoenbrun

Conjuring the Modern in Africa: Durability and Rupture in Histories of Public Healing between the Great Lakes of East Africa

The valuable emphasis on modernity in colonial and postcolonial African studies has profoundly divided precolonial African history from what comes after. But the depth and complexity of African aspirations for moral community and the forms of collective action they inspire, often in the face of severe material constraints, exceed the explanatory power of narratives of modernity oriented toward the history of capital, colony, and commerce. Long-term regional histories of durable bundles of meaning and practice grounded in Africa address these matters in part by working across tight spaces of ethnicity and beyond shallow chronologies. In particular, a history of public healing reveals compelling notions of public health and forms of power that cut across the colonial period but were transformed by colonialism. Public healing has wrestled with shifting boundaries between a porous social body's moral communities and the starker outline of an embodied, autonomous individual. Modes of power and authority central to politics and to healing practices, public or private, have moved uneasily but productively against each other over the last millennium, as agricultural systems changed, as centralized states formed, and as commodified economies grew. Over the last century, they have moved against the forms of power and authority embodied in a colonial state or in biomedicine. In the context of public healing between the African Great Lakes, the historical complexity of relations between these entangled aspects of life reveals a heterotemporal modern Africa beyond the hybrid or the alternative forms of modernity so prevalent in the literature.

Since the 1990s, work in African colonial history has emphasized African appropriations of European forms of knowledge and practice in a single field of culture inflected by political economy. These ethnographically dense explorations of what has been called "the colonial situation" push beyond a paradigm of "encounter" into a contingent history of colonial and postcolonial modernity. One scholar, Nancy Rose Hunt, finds valuable sources for her analytical categories in a precolonial Central African history of bodies, gender, aspiration, and mobility by focusing on "colonial middle figures" such as nurses.¹ The historian Florence Bernault argues that a single field of violent struggle over power enveloped Europeans and Africans in a common set of practices related to the body, some of which have roots in precolonial African histories.² Jean Comaroff and John Comaroff also focus on middle figures and power, but precolonial historical trajectories play an inconsistent role in their analysis. Sometimes, they condense them in the moment just before Europeans arrived in the interior of southern Africa.³ Other times, they speak of a steadfast continuity in the trajectory of African practices: "the will to assimilate the knowledges of alien cultures was inherent in the epistemes and ethnopractice of these peoples even before the arrival of the Europeans, and has remained so ever since." Mostly, the Comaroffs explore the ways in which precolonial African cultural practices were utterly transformed by encounters with Europeans.⁴ Must their valuable focus on transformation sustain an analytical divide between the "much more" of what came before and all that did get drawn into "the colonial encounter"? If some of the terms on which the colonial transformations unfolded have their roots in what came before, their earlier histories need study.⁵

Steven Feierman doubts that framing colonial history in terms of appropriation opens up "how patterns of action and forms of signifying practice within African societies came to be understood as fragmented and partial" and reveals "the process by which important cultural domains came to disappear."⁶ He worries that the specificities of colonial-cultural mixtures tend to make historical sense in terms of layered narratives that "originate in Europe," particularly the narratives of capitalism and of Protestantism and the implicit, general sense of their historical relations to each other. They offer "a certain coherence" of historical imagination concerning the African past, even though "we can expect" each of these stories "to exist in creative tension with larger historical narratives ... the central question is, which larger narratives?"⁷ The missing narratives are long regional histories of Africa, flawed and compressed and dependent upon concepts whose explanatory status must be contested.⁸ This regional history "has a necessary role to play if historical knowledge in the aggregate is not to do violence to understandings which grow out of microhistorical study" of what Africans did and what they thought they were doing when they began to engage the currents of capitalism.⁹

One of these creative tensions lies in recognizing that something is entangled with the modernity of the moral communities through which Africans pursue their aspirations and address issues of need.¹⁰ This something else, which I hesitate to name, comes to light in regional historical processes with African roots. Dipesh Chakrabarty has argued that European narratives of capital and colony, and the European thought they fostered, are "both indispensable and inadequate in helping us to think through the various life practices that constitute the political and the historical."¹¹ Chakrabarty writes about India, but the provision here of a historical narrative of public healing engages his conundrum by serving as the something else necessary to making new sense of East African modernity. Africanist historiography has long been concerned with this new perspective. But even a brief tour of it reveals that the boundary between the modern and this something else can easily be overdrawn.

The colonial experience created Africanists by rendering Africans historyless, traditional people.¹² Tradition was the shadowy space that Europeans and Africans used to figure their emergence as "moderns" in colonial settings.¹³ Africanism and Africanists—especially historians—engaged this divide by demonstrating that African pasts were full of change and dynamism.¹⁴ Two branches of historical scholarship on Africa grow from these demonstrations. One studies the periods just before and including colonial conquest, colonial rule, independence, and postcolonial issues. The second engages the sixteenth century to the nineteenth century, when African economies and societies were drawn into a world of slavery, mercantilism, and industrialization. To some extent, histories of Islam in sub-Saharan Africa cut across these themes, but even they keep largely to the centuries after 1500.¹⁵ The work of two leading scholars in African history—Jan Vansina and Steven Feierman—reveals the value of linking these branches of Africa scholarship both to each other and to the history of Africa before the sixteenth century.¹⁶

Jan Vansina explores the nature of tradition and assesses the conditions for its survival through transformation in his book Paths in the Rainforests.¹⁷ He narrates three thousand years of historical change in what he calls the Equatorial African political tradition, a version of tradition that emphasizes both enduring continuities in its institutional and intellectual constituents and continuous change in their interrelationships.¹⁸ The book refutes persistent suggestions that Equatorial African societies before 1500 were changeless, opening the way for the comparative study of traditions the world over.

But Vansina also argues that between the 1880s and the 1920s, the violence of colonial conquest extinguished the equatorial African tradition.

As a result [of conquest], the peoples of the rainforests began first to doubt their own legacies and then to adopt portions of the foreign heritage. But they clung to their own languages and to much of the older cognitive content carried by them ... , striving for a new synthesis which could not be achieved as long as freedom of action was denied them.¹⁹

The particularly violent conquest of the Inner Congo Basin destroyed the premise of autonomy at the core of an Equatorial African political tradition. But loss and extermination—familiar themes in Central African history—are not the only fates for "tradition" after colonial conquest.²⁰

In his book Peasant Intellectuals, Steven Feierman argues that fragmentation lies at the creative heart of a tradition. Feierman emphasizes the role of discourse in the creative uses of tradition by its makers. He understood this creativity as acts of selecting and composing arrays of cultural material to meet particular challenges. The creative and selective use of tradition, shaped by an aggressive colonial rule, reveals how certain things, such as forms of political language, remain in play across the divides between precolonial, colonial, and postcolonial experience.

When people select a particular form of discourse, when they shape a political argument in a particular way, this is by no means a passive act. The social analysis of peasant discourse in this book will show that long-term continuities in political language are the outcome of radical social change and of struggle within peasant society.²¹

The appeal of Feierman's conclusion lies in seeing precolonial cultural legacies as provisional and under construction by historians, but as forming part of Africans' "resources and modes of empowerment in seeking to act effectively in and upon a changing world."²² They open up the logics of African ideas and actions in colonial and postcolonial settings as part of an ongoing program of effective moral action, the work of people who use bundles of durable language to confront circumstances of real material struggle, shaped by both African and external historical processes.²³ As others, including Vansina and Feierman, have argued, the ruptures of social and political institutions and of intellectual traditions in Africa during the nineteenth century were not the first radical transformations shaped by African bricolage and durability.²⁴ The multiple transformations of meaning and practice in public healing have moral and collective legacies that are especially clear in East Africa today.

In 1996, Geoffrey Kamali, a reporter for Uganda's government-sponsored newspaper, the New Vision, wrote a story about visiting the shrines of ancestral spirits, based on a conversation with an anonymous woman. She told him that she went out in the middle of the night in a fleet of seven taxis filled with people following someone called Jjajjà ("grandmother, grandfather, ancestor, founder" in Luganda, a major language in Uganda), who knew how spirits behaved and what sorts of things their would-be supplicants should and should not do. The taxis carried mostly women. They sought a shrine on a hilltop outside Kampala, Uganda, famous for its three nearby caves. They carried cash and coffee berries to give to the omusámbwa of that place, the spirit of the caves. When they reached the shrine, they were told that they could not wear their overcoats because they should not imitate the spirit, who liked to wear an overcoat. They were told that they would see the omusámbwa in a cave, after they left their coffee berries and cash as offerings. Only one visitor, a man, claimed to have seen the spirit of the place, a man-like, very tall omusámbwa, waving a burning tree in his hands. After this sighting, a bonfire was lit around which people danced and drummed, asking for money, a better job, education, and fertility. Some visitors became possessed by spirits. Jjajjà, the grandparent–ancestor guide, moved through the huge fire without being burned. After this, Jjajjà got everyone back in the taxis and conducted them to the shores of Lake Victoria. At her house, she tattooed the visitors' right arms, asked them to confess their bad deeds, and gave each person a number of coffee berries to swallow.²⁵

The people in this scene want more money and better skills in order to compete in a tight labor market and to have more choices as consumers. Scholars commonly analyze these desires in terms borrowed from the histories of economies and cultures and religions with deep and broad roots in Europe or North America.²⁶ But grasping how the supplicants conceptualize and go about their business—the things and practices they use and the ideas with which they debate possibilities and desires—exceeds the explanatory power of the history of the impact of capitalism, even while they cannot be explained without it.²⁷ The logics of collective action and moral community at work in Kamali's story lie beyond an African history of capitalism at the same time that they helped shape the transformative experiences of capitalism in Africa.²⁸ They also lie beyond the explanatory power of "a generic colonialism" that has been "given the decisive role in shaping a postcolonial moment."²⁹ Regional histories of healing practice broaden explanations of this public business of seeking burning elders by shifting the burden of analysis from a focus on capitalism, colonialism, and religions to the historical study of kinship, royalty, spirits, and fertility.³⁰ These themes help historians "examine how indigenous peoples struggle to integrate their experience of the world system in something that is logically and ontologically more inclusive: their own system of the world."³¹

Map 1. The Great Lakes Region of East Africa. Map made by Tom O'Connell, Digital Media Services, Northwestern University.

The shape of such a system hides in plain sight in Kamali's scene as people pursue their aspirations with techniques and ideas of both great antiquity and more recent vintage.³² The figure of the unburned ancestor (called Jjajjà here) turns up in oral narratives far to the southwest, on the border with Rwanda, before, during, and after the course of colonial conquests there.³³ The category of omusámbwa, a territorial spirit, occurs in societies around the entire circumference of Lake Victoria and has a life in the region many centuries old.³⁴ Coffee berries have been used in contracting blood friendships, a common way for people in the Great Lakes region to build ties that supplement kinship.³⁵ Dancing, drumming, and possession by disembodied spiritual personae are found together across Bantu-speaking Africa, from Angola to Zimbabwe, and have a history several millennia in depth.³⁶ Cash and coats in the region are deeply intertwined with the slaving and violence and mission work of the nineteenth century.³⁷ Kamali's scene contains a historical iconography of discourse and practice that conjures precolonial, colonial, and postcolonial contexts in multiple ways. This bricolage blurs the "radical disjuncture" between "an inferior past" and "a superior future" implicit in a modern temporal ideology of aspiration.³⁸ These entangled times—or heterotemporalities—are central to Chakrabarty's call to "contemplate the necessarily fragmentary histories of human belonging that never constitute a one or a whole."³⁹ If one of the effects of modernity on "traditional" worlds is to bring history to them, heterotemporalities return the gift by pushing beyond "the empty and homogeneous time" of modernity.⁴⁰

Existentially poetic stories of the fragmentary are also stories of struggle. Flawed and freighted, the history of public healing emphasizes the discourses and practices that people have used in "the boundary-crossing struggle over the conceptual and moral bases of political and social organization." It reveals in the depth of these discourses and practices a thread of struggle that Frederick Cooper says is often "lost in opposing European, capitalist, imperialist `modernity' to `alternative modernities' or a space of the nonmodern." As Cooper goes on to argue, the power to create and establish claims "and to alter definitions of what is a debatable issue and what is not" is unequal in any historical context; it is therefore crucial "to keep one's focus on how such concepts were used in historical situations."⁴¹ Constituting the sources to meet these challenges—and asking how the sources have been constituted—is a necessary first step in exploring the multiple temporalities recognizable today in the moral community and collective action at the core of public healing in Africa.

Such histories have unusual textures and contents because they rely on oral, linguistic, ethnographic, archaeological, and environmental evidence and because they work with unconventional units of historical agency and subjectivity. Their time frames are broad, covering centuries and even millennia. Archaeological sites, speech areas, or vegetation zones define different spatial units of analysis.⁴²

Environmental studies of the Great Lakes region focus on climatic and vegetation change, seeking to distinguish human from other causes.⁴³ Oral traditions often mention droughts and famines, supporting inferences about the strains of political economic change as well as the conditions that valued such historical memories.⁴⁴ In the second half of the nineteenth century, as the suite of supporting sources grows denser in number and kind, the conceptual complexity of environmental histories of the region grows increasingly rich. The apparently straightforward nature of environmental evidence can then be read through, as well as read into, regional histories of demography and health.⁴⁵

Archaeologists study material culture, spatial patterns, and technological change. The patterning and variability of material culture change over time, revealing much about demographic and social processes, but cultural meanings are inferred by analogy from better-documented contexts, often at a great remove in time from the archaeological context itself.⁴⁶ In considering the practice of analogy, Alison Wylie argues that information exchanges may move in both ways; the archaeological record may subvert conventional wisdoms used to construct ethnographies and historical documents.⁴⁷ In the Great Lakes region, archaeologists have excavated a number of earthworks sites and explored their associations with oral traditions about departed royal dynasties.⁴⁸ Some scholars have situated their interpretations of the function of these sites in a hermeneutics of power and authority drawn from the oral sources, while the dating and structure of the sites suggest older, not necessarily royal associations.⁴⁹

Oral traditions usher in actors as composite figures of persons or groups.⁵⁰ This is especially true in dynastic histories purporting to speak of very early periods in narratives of the actions of royals and nobles, and perhaps of mediums, but not of everyone else.⁵¹ In some parts of the Lakes region, Feierman has observed that the unfolding of genealogies dominates such dynastic traditions, and that genealogies express a sort of time weighted strongly in the direction of masculine forms of authority.⁵² Such traditions highlight change and continuity in particular formulations of the masculine; other historical tales reveal even more about gender dynamics.⁵³ But separating heard from read and uttered from written messages and testimony is often misleading, and not infrequently just plain impossible. The contents of a written source draw on other types of sources, and any source of one type or another can be put to use in other contexts.

Reconstructed language histories and comparative ethnographic evidence support much of the narrative offered below. One begins by classifying related languages, determining the nature of those relationships, proposing sequences of linguistic divergences, and establishing the basic similarities and differences in vocabulary, phonology, morphology, and tone systems.⁵⁴ The goal is to reconstruct the vocabulary of earlier "proto-languages."⁵⁵ A proto-language is a historical archive of the continuities and innovations in words and meanings, constituted by their transmission across the generations and representing the durability of speech communities over time.⁵⁶

Hypothetical earlier meanings may be built on the nature of their distributions in the region's languages and on a set of assumptions about the direction of semantic change. If a word with the same form and the same or similar meaning occurs in a set of contemporary languages "known to be related, it is most probable that all variations were inherited from their common ancestral language."⁵⁷ Words such as mugàngà, "healer, doctor, operator of medicine objects," whose meanings are very widely distributed, reflect this sort of situation. (See Semantogram 1.)⁵⁸ If the distribution of the variations in meaning that constitute a widening or a narrowing of a word's semantic field confines itself to a subgroup of a set of related languages, then it is most likely that those variations emerged when the language ancestral to that subgroup was spoken.⁵⁹

SEMANTOGRAM 1. Social Sources of Health and Illness [Read from top to bottom in order to go from the past to the present and perceive meaning retentions and shifts]
	-zimu* n. 1/21 From -dįma* 'to extinguish'	-bándwa* v. From -bánda* `to split' (tr.) [plus passive suffix]	-sámbwa* n. 3/4, 5/6 From -samba* `judge, render justice, bless' [plus agentive noun suffix]	-cwezi* n. 1/2 From -cwera `to spit'	*lubaale n. 11/2 From -baale* `stone'
Proto-Bantu ↓ ↓ ↓ ↓	'Ghost of long-departed person; form spirit might take'	'Split or cleave'[not in passive]	'Be judged, have justice rendered, be blessed'[with passive suffix]	Not present	Not present
Proto-Great Lakes Bantu ↓ ↓ ↓ ↓	Retained; may become a -sámbwa* In class 17 `land of the dead'	'Be seized in head by spirit' (*-zimu, -sambwa,* others?) [plus passive suffix] `Be consecrated to spirit'; n. `spirit type; medium'	'Lineage spirits, the territory they control, form spirit might take' (python, whirlwind, stone, tree, stream, spring, pool of water)	Not present	Not present
Proto-West Nyanza Bantu ↓ ↓ ↓ ↓	Retained	Retained	Retained; but included named, portable spirits (such as Mukasa) controlled genres of experience and environmental categories	Not present	Not present
Proto-North Nyanza Bantu	Retained; but seen as harmful?	'Be possessed by lubaale spirit or by muzimu, be consecrated to such spirits'	Retained	Not present	A sort of mbándwa spirit or its medium; later, Ganda `national' deities
Proto-Rutara Bantu [Descended from Proto-West Nyanza Bantu]	Retained; but seen as harmful?	Retained plus `be possessed by -cwezi* spirit' Forms: mbándwa eziragura ('black spirits'), mbándwa ezeera ('white spirits')	Retained only in languages along Lake Victoria's coast (Haya dialects, Zinza)	'Spirit with no heirs, its medium' From mbándwa ezeera category; later included named spirits of departed leaders	Not present
Proto-Western Lakes Bantu [Descended from Proto-Great Lakes Bantu]	Meaning from Great Lakes Bantu retained; no relation to -sámbwa*	Meaning from Great Lakes Bantu retained	Not present	Not present	Not present
Proto-West Highlands Bantu	Retained	'Be consecrated to special cult (Ryángombe, Kiranga, Nyabingi)'	In Ha (Heru) refers to Zinza spirits	Not present	Not present
1 The numbers following the root refer to a system of classifying noun prefixes denoting singular and plural forms; see Francis Katamba, "Bantu Nominal Morphology," in Nurse and Philippson, Bantu Languages, 104.

The plausibility of these inferences depends on the strength of the genetic classification of the languages included in the comparison; strong overall classifications produce clear subgroups of languages within them.⁶⁰ The genetic classification of Great Lakes Bantu—the earliest genetic unit in play here—is fairly secure. Therefore, the historical relatedness of the major subgroups at the core of the early parts of the history of healing practice is well-attested, generating confidence in the shape of reconstructed words.⁶¹ (See Figure 1.)

Figure 1. Outline Classification of Great Lakes Bantu1


Great Lakes Bantu
	1. Western Lakes
		A. Rwenzori
			Koonzo, Nande
		B. Pre-Kabwari
			Kabwari
		C. Kivu
			Forest
				Tembo, Nyanga, Shi, Hunde, Havu, Fuliiru, Vira
			West Highlands
				Rwanda, Rundi, Ha, Hangaza, Shuubi, Vinza
	2. West Nyanza
		A. North Nyanza
			Ganda, Soga, Gwere, Shana
		B. Rutara
			North Rutara
				Nyoro, Tooro, Nkore, Kiga
			South Rutara
				Haya, Nyambo
			Kerebe
				Zinza
	3. Pre-Gungu
		Gungu
	4. East Nyanza
		Mara
			South Mara
				Zanaki, Ngoreme, Nata, Shashi, Zu
			North Mara
				Gusįį, Kuria, Simbete
		Suguti
			Jita, Ruri, Regi, Kwaya
	5. Greater Luhyia
		Southern Luhyia
			Takho
		Central Luhyia
			North Luhyia
				South Masaaba
				Dadiri
			Nyole
			Saamya

1Many dialects in some of these languages exist or existed. See Schoenbrun, "Great Lakes Bantu," fn. 1.

Their etymologies constitute a major source of historical evidence. Etymologies are based on common derivational processes in Bantu languages and on the distributions of forms and meanings.⁶² The verb kusamba occurs in regularly corresponding form carrying the same meanings—"judge" or "bless"—in many, many Bantu languages. It arguably formed part of the proto-Bantu vocabulary. A passive form of the verb, kusambwa, meaning "to be judged, to be blessed," conforms to the same criteria. But a noun, musámbwa, referring to a territorial spirit and the physical form such a spirit might take, occurs as a regularly corresponding form and meaning in a limited number of languages belonging to constituent branches of Great Lakes Bantu. The noun and its new meanings were apparently invented by people who spoke Great Lakes Bantu.

Semantic histories reconstructed in this manner lack the contextual nuances that listeners and speakers draw on to communicate. The meanings attached to reconstructed vocabulary are also culturally flat, because they collapse competing forms of meaning. But they are not agentless abstractions, because the distinctions sustaining meaning result from action and reflection in specific contexts.⁶³ In other words, the abstractions of meaning, given in a "gloss," represent the durable intellectual and practical contents of the social worlds inside of which people acted. We have no acts, no disagreements over values and strategies. Not a single utterance can be reconstructed. Instead, the results of past actions, disagreements, and speech take the form of inherited and innovated words and meanings. The stability, narrowing, and broadening of fields of meaning referred to by the terms in the semantograms reflect people's work in achieving continuity and change.⁶⁴

This rich pool of sources tries conventions of historical narrative. Dating and region are vague, gaps in content are common, and chronologies are imprecise. While such narratives of the longue durée are flawed and provisional, they are also clearly amenable to exploring durability and transformation in fields such as collective well-being and moral community. These categories, and some of the settings in which they have force, push past the binary of the antique and the modern in analyzing current struggles over social justice.

Collective well-being and moral community lie at the heart of scholarship on what Mikael Karlström calls the dystopian spirit of witchcraft and the critique of contemporary capitalist relations and political life that it is said to express.⁶⁵ Karlström argues that the "moral community" brought to life in monarchy and in rituals of social reproduction represents the dynamic obverse of the revenge and greed at the heart of witchcraft in Buganda, a monarchy on the north shore of Lake Victoria. The power of witchcraft discourses, Karlström argues, seems "to lie overwhelmingly in their capacity to objectify the perils of illicit power and the antisocial dangers of exploitative accumulation and self-interested consumption."⁶⁶ If witchcraft is "the force that both generates and feeds upon violations of the fundamental norms of kinship solidarity"—kinship, social prosperity, reciprocity, and hospitality—then "forms of moral community and their modes and nodes of reproduction" should be explored as arenas that Ganda people use to keep their aspirations in play in a world stacked against them.⁶⁷ Ganda use their kingdom as one means to think about moral communities.

Buganda is a famous monarchy on the north shore of Lake Victoria. It is commonly held to have been created when a first king integrated clans into a project of building wealth and order through overlapping networks of reciprocal obligation with himself, a Queen Mother, a Queen Sister, and their courts at the center.⁶⁸ When they talk today about their monarchy, Ganda people insist that moral communities exist because a king exists: "without a king, no clans or lineages, no history or meaning, no morality or culture" would exist.⁶⁹ A complex history informs this compact equation and lies beyond our scope here.⁷⁰ But the claim points to how people draw on indigenous social and historical imaginaries while appropriating Western models of social, religious, and economic life in the name of "tradition."⁷¹

Karlström shows how, in the 1920s, Ganda articulated "a hybrid sociotemporal consciousness" through a revaluation of custom aimed at "securing a moral collectivity" against the fractious threats of antisocial individualisms. In the 1990s, this gave way to a cultural revivalism of their monarchy aimed at protecting a moral collectivity from an unstable postcolonial state.⁷² Both of these efforts rested on redefining and reusing existing languages of moral community. However, Kamali's scene—which unfolds inside Buganda—does not fit with Karlström's picture of a Ganda cultural royalism "concerned with constituting visible and circumscribed, hence legitimate, relations of authority and hierarchy"; nor does it fit with conventional views which hold that witchcraft practice "expresses and generates intergenerational antagonism."⁷³ People worry about power and wealth—but not as something "socially divisive, destructive, secretive, parasitic." They articulate individual needs and desires, but they do so in a public setting not clearly circumscribed by hierarchy and authority. The practices of social reproduction in Kamali's scene refer literally to a topos of "moral sociality": the omusámbwa spirit and its abode. The sociality of the scene is unstable, and the authority of Jjajjà is fleeting. Studying the regional historical trajectories of the different ideas that people used to think about health and to practice healing repositions the contents of moral communities, moral behavior, and the changing institutional and conceptual sources for these vectors of action in Africa. To appreciate why this is so, one must be familiar with the social basis of health and healing practices in African history. Only then can one understand how the history of public healing is central to larger themes in the history of the Great Lakes region.

Health in African history implicates histories of the environment, of the state, of gender; it is a social history. Healing in African history implicates other histories, too—of morality, of the body, of the person, of relations between life and death, of notions of efficacy and capacity. They are histories of practical reason—in intimate as well as in public life—as much as they are histories of the forces that cause illness and sustain wellness.⁷⁴ African histories of healing intersect with these larger narratives, but they must grapple with concepts of causality not easily translated across cultures and forms of action greatly concerned with "the social embeddedness" of suffering and misfortune.⁷⁵

Scholarship on precolonial healing in eastern Africa has concentrated on the nexus of causation binding therapeutic approaches to illness. In Feierman's formulation, diagnoses move between two relatively stable categories: illnesses of God and illnesses of people. Illnesses of God just happen, but illnesses of people are caused by human action, especially the vengeful and antisocial actions of people labeled "sorcerers" or "witches." For example, people in many parts of eastern and southern Africa often explained droughts as the work of an enemy; "in a sense," a drought was "an illness of humanity raised to a public level."⁷⁶ Yet, not all droughts were understood in this way. Some just happened, especially those which exceeded the territorial reach of a given healer. Illnesses of people precipitated actions involving "counter-sorcery" by a healer or a rainmaker greater than the one creating the illness or drought. By this causal logic, people approach treatment through trial and error rather than through a differential diagnosis.⁷⁷

These formulations of causality defined the scope of public health practices. Scholars distinguish offerings and sacrifices to the ancestral figures of a particular group of kin from propitiating figures linked to larger communities with spheres of efficacy beyond those of families or close kin.⁷⁸ The powers of kings, court ritualists, chiefs, certain mediums, and rainmakers—common actors in the domain of public health—are often understood to work in this expansive scope. Spirit possession activities cross these boundaries; a family spirit may be the source of public dancing in which participation was elective.⁷⁹ Epidemic crises precipitated efforts to control people's movements and quotidian activities, although these issues have been little studied before the nineteenth century.⁸⁰

Healers and patients relied on medicines, "substances with powers to transform bodies," in order to achieve health.⁸¹ A medicine's transformative capacities came from its substance, from speech, and from the individuals and collectivities who made and used it. Some medicines—such as those given for upset stomach—transformed bodies solely because of their substance. But other medicines—including those used to keep thieves out of a field of ripe crops or those used to keep hail from falling—transformed material realities (not just bodies) because of their substance, because of their activation by the speech of a powerful person or persons, and because of other actions of persons and collectivities.⁸² Substances used in complex, protective "medicines" (often called "charms" or "amulets" in European languages) were often central to "completing the metonymic chain from the original power source to its specific beneficiary."⁸³ Understanding how medicines work redraws the boundaries between materiality, speech, the existential, and the social, depending on the nature of the outcome that supplicants seek to bring about or to forestall.

The illnesses on which medicines were designed to act were matters of experience and reflection as well as of biology. They were a "syndrome of experience" and not a "mirror of nature."⁸⁴ As the historian Julie Livingston has argued, this means that diagnosis and treatment tended to reinforce "the overarching unity of the natural, cosmological, and social realms—all of which were in flux."⁸⁵ Even with the richest imaginable collection of historical evidence, it is difficult to untangle the particular changes in the relations between these realms. But because the language of illness was socially constructed, authorized, and contested, historians of precolonial Africa can probe aspects of the role of language in these processes when they compare the evidence of vocabulary in reconstructing the shape of interlocking ideas, practices, and things that "run together for members of a society."⁸⁶

The institution called ngòmà exemplifies some of these issues. The term ngòmà means "drum" in Bantu languages from Cameroon to Namibia and from Kenya to Zimbabwe. It also means a sort of musical performance in a long swath of languages east of the Central African forest zone, reaching to South Africa.⁸⁷Ngòmà is a set of public healing practices, materials, and settings that people use to pursue health, wealth, and protection of entire communities and of evanescent collectivities from natural or spiritual social dangers, and to commemorate departed persons. Through ngòmà, people make sense of misfortune, manifested in disease or symptoms of disease, attributed to various sorts of disembodied spiritual beings. Treatment provides them with a name for their misfortune and incorporates sufferers into a larger community of afflicted persons.⁸⁸

In general, ngòmà involves one or more sufferers, one or more healers, a group of dancers or singers and musicians, and their musical instruments (drums and rattles, and often whistles, clappers, zithers, and harps). Ngòmà involves invoking ancestral, nature, or territorial spirits, sometimes by a medium, who may or may not belong to the group of healers. The kinesthetics of drumming and dancing centers public performance and staging in healing through ngòmà.⁸⁹ All across the territory of ngòmà, the sufferer sings to the assembled group of supporters and healers, who sing back. When the public healing of drumming and possession dances occurs beyond royal, colonial, or police surveillance, it may embody a refusal of the commonsense view that force leads to control.⁹⁰ The therapeutic core of the institution might be understood as leading to the "consolidation of substantial resources, material and human," and as aiding "the long-term reordering of institutions of redress, economic redistribution, and ideological change."⁹¹ The flexibility of these healing practices in treating afflictions of radically different scales and phenomena follows from ngòmà's modular form, a form that facilitates innovation.⁹²

Healing practices in East Africa change as people assess the efficacy of treatments in terms of practical reason and moral principles. People in Uganda embraced injections because they perceived the effectiveness of the medicines delivered by that technique, and "efficacy provides evidence that changes the culture of health."⁹³ When people say that an elder's capacity to curse her juniors has caused an illness, they represent both the cause and the effect of cursing in terms of social relations and moral power. Her moral capacities contrast with the vengeful, greedy desires behind the immorality of witches. The contents of morality and immorality change as environmental, agricultural, and political processes engender struggles and debates over them, closing down and opening up what it is possible to argue about and what it is possible to envision as "a morally better or worse world than the one in which" people lived.⁹⁴ The substance of an African regional history of durability and rupture in the tense relation between existential uncertainty, the challenges of providing care, and healers' attempts to manipulate reality raises questions about how that history reconfigures approaches to more recent moralities and modernities.

What practices run through the history of public healing between the Great Lakes?⁹⁵ Spirit possession and mediumship have been in play from earliest times, as revealed by the historical development of the meanings attached to the term kubándwa. (See Semantogram 1.) Today, the verb means "to be possessed by or consecrated to a spirit," in a discontinuous distribution that includes languages in all branches of Great Lakes Bantu, except East Nyanza.⁹⁶ (See Figure 1.) The meaning represents an innovation from the verb kubánda, which means "to press down or knock down" in a still wider contemporary distribution.⁹⁷ Great Lakes Bantu–speaking healers and patients likened the experience of being possessed by a spirit to the feeling of being overwhelmed or knocked down. They made a noun from this verb to name the spirit who did the possessing and the medium who had been possessed. Travelers' accounts from the nineteenth century describe spirit possession as tackling personal health concerns, concerns for the health of entire communities (defined by a particular locality or set of relatives), and concerns with the health of entire territories (mainly with ensuring or restoring their fecundity).⁹⁸

The specialists mentioned in Semantogram 2 integrated the social basis of knowledge relevant to health and healing with the power of speech, most clearly in public forms of healing such as possession (kubándwa), offering sacrifices (kutámba), and divination (kulàgula).⁹⁹ Descriptions of these forms of healing in the late nineteenth and early twentieth centuries repeatedly mention the presence of numerous participants, patients, their families, onlookers, mediums, and priests and doctor-diviners themselves.¹⁰⁰ Spirit possession and propitiation, often organized by specialists, created large evanescent publics.

SEMANTOGRAM 2. Practitioners and Practices [Read from top to bottom in order to go from the past toward the present and perceive meaning retentions and shifts]
	-làgula* v.; -làgudį* n. From -dàg-* `teach; promise' plus reversive suffix, giving v. `exchange teachings or promises'	-kúmú* n. Etymology uncertain	-gàngà* n. From -gàngà* v. `to tie up'	-támba* v. From *-támba `to set a trap' or 'to walk, travel'
Proto-Bantu ↓ ↓ ↓ ↓	Not present; in Proto-Mashariki: v. `Divine, prophecy'; n. `divination specialist'	'Leader, person of wealth'	Not present? In Proto-Mashariki: `healer, doctor'	Not present; in Proto-Mashariki: `offer, sacrifice (by slaughtering) in order to heal'
Proto-Great Lakes Bantu ↓	Retained	'Doctor-diviner, general practitioner'	Retained	Retained
Proto-West Nyanza Bantu ↓	Retained	Retained included diagnosing spirit possession	Retained	Retained
Proto-North Nyanza [Descended from Proto-West Nyanza Bantu]	Retained	Retained in Proto-West Nyanza meaning	Retained	Retained, in Proto-West Nyanza meaning. Later, may have included human sacrifice in Buganda
Proto-Rutara Bantu [Descended from Proto-West Nyanza Bantu]	Retained; now included mbándwa eziragura spirits	Retained in Proto-West Nyanza meaning; later they might be present at royal courts	Retained	Retained in Proto-West Nyanza meaning
Proto-Western Lakes Bantu ↓	Retained	Retained in Proto-Great Lakes Bantu meaning	Retained; added meaning in class 14 `good health'	Retained
Proto-West Highlands Bantu	Retained	Retained in Proto-Great Lakes Bantu meaning	Not present; replaced by mufúmú n.	Retained; now included dancing

Healing practice relied on powerful speech, clearly expressed in the meaning of "to divine" or "divination." The West Nyanza and Western Lakes word for these acts was derived from a verb that meant "to promise" or "to teach." By adding a reversive suffix to that verb, speakers of these languages expressed the act of promising or teaching things and ideas to or for others. This morphological innovation is widely distributed in eastern Bantu languages and is quite old. Its retention in proto–Great Lakes Bantu reflects key aspects of being a healing specialist between the Great Lakes in the latter centuries b.c.e., when Great Lakes Bantu existed as a speech community. Diviners were master speakers, and they were master listeners. Their ability to "divine" the cause of an individual's, a collective's, or a territory's misfortune relied on eliciting and reframing information.

Great Lakes peoples thought about health and prosperity together. Their diviner-healers, the bafúmú, possessed a variety of powers, among them the power to heal or cure, kukíla, which also meant, for Lakes peoples, "to prosper." They used an even older word, mugàngà, to name doctors.¹⁰¹ But they understood a mugàngà to be particularly gifted in one or another area of the larger territory of healing practice, including divination.

The spirits, specialists, medicines (see Semantogram 3), or concepts that healers and patients used over the last fifteen hundred years exceed the groupings given in the semantograms, but they represent a basic, generative architecture within which people innovated and revised healing practices during various periods of rupture in Great Lakes history. These terms, which have been confidently reconstructed as parts of the vocabulary of West Nyanza and Western Lakes speech communities, formed the practical and theoretical material with which later speech communities created new modes of public healing. They used other words to name grain crops, domestic animals, and farming techniques either directly or indirectly attested in the region's archaeological and paleoecological records. Both glottochronological and radiocarbon dating methods point to the middle of the first millennium c.e. as the time during which these speech communities of farmers and herders (and their grains and cattle) and public healers existed.¹⁰²

SEMANTOGRAM 3. Medicines and Techniques [Read from top to bottom in order to go from the past to the present and to perceive meaning retentions and shifts]
	-komero* n. 7/8. From v. -komera* `be strong, alive' < v. -kóma* `become finished' + prep. suff. > `realize potential'1	-sango* n. various classes. From v. -sánga* `find, discover'	-gica/-gico n. 9/10. From v. *-kita `to do'	-ti n. 3/4, 14 From n. -ti `tree, shrub'	-sigo* n. 1, 7, 14 From v. -siga* `leave behind, abandon'
Proto-Bantu ↓ ↓ ↓ ↓	?	Not present; in Mashariki as `charm, consecrated medicine'	'Charm, blessing' (enabled by speech)	'Medicine' Often in class 14 as `essence' of trees or shrubs'	? `Offering, vow, sacrifice'
Proto-Great Lakes Bantu ↓ ↓ ↓	'Medicine; Efficacious thing when activated by speech'	'Charm; Efficacious thing when activated by speech or song'	Retained	Retained	Retained
Proto-West Nyanza Bantu ↓	Retained	Not present	Retained	Retained	Retained
Proto-North Nyanza Bantu [Descended from Proto-West Nyanza Bantu]	Not present	Not present	Retained	Retained	Not present
Proto-Rutara Bantu [Descended from Proto-West Nyanza Bantu]	Retained	Not present	Retained; now included specific materials (wood and ntembe seeds) and use by diviners and Mukasa's mediums or priests	Retained	Retained
Proto-Western Lakes Bantu ↓ ↓	Retained	Retained in Great Lakes Bantu meaning	Retained in general meaning `charm, blessing'	Retained	Retained
Proto-West Highlands Bantu [Descended from Proto-Western Lakes Bantu]	Retained	Retained	Retained	Retained	Retained; comes to mean `sort of territorial nature spirit' after c. 1500
1Could also be from a verb meaning `to hit, strike' (plus prepositional suffix), indicating how words spoken over medicine release or open its capacity to heal.

The semantic histories of words for medicines, spirits, specialists, and techniques underscore the close relations between ideas of health and conditions of prosperity, implying that healing practices existed in creative tension with social organization and political culture. These two large domains of life disrupted or ensured conditions of prosperity, rendering the practical aspirations of persons, communities, and leaders in social life the proper territory of healing. Healing institutions were designed by their practitioners to cure illness, end famine, remove epidemics, and so on. They were about action, the capacity for action, and the particular moral and social conditions that shaped action.

In the mid-first millennium of the Common Era, West Nyanza, Greater Luhyia, and East Nyanza societies, living around and near to the shores of Lake Victoria, drew on the concept of ancestral ghosts (the bazimu) to invent a new category of spirits that could reside in natural locales, such as caves, springs, lakes, or rivers, and whose range of efficacy had clear territorial dimensions. They worked out the meanings and capacities of these novel territorial nature spirits during a time of agricultural specialization. Productive pastoralism, grain and banana farming, began to generate conditions that induced leaders to create the concept of primacy in a region and to develop patronage systems designed to compose communities of diverse followers while at the same time controlling access to productive forms of wealth such as cattle herds and banana gardens, through exclusionary inheritance systems based on descent.¹⁰³

People thought of territorial spirits, which they called misámbwa or masámbwa (sg. musámbwa), as ancestors of the first groups or lineages in a territory.¹⁰⁴ They were implicated in the fecundity of a territory and the prosperity of its residents. This range of efficacy differed from that of ancestral ghosts, whose interests extended only to members of a particular family or of an extended lineage. Firstcomer groups drew on the knowledge that bazimu could return from okuzimu (lit. "where ancestral ghosts exist") or Ghostland as animals or snakes in claiming that their ghosts returned for the good of communities larger than a single lineage or neighborhood. Their ghosts, now called misámbwa, resided in the territory of the firstcomer lineage, and caring for them ensured fecundity and prosperity in their territory and fertility for all its residents, be they members of the firstcomer lineage or not. Twentieth-century sources report that rocky hills, large trees, or springs were places where people encountered such spirits.

Late in the first millennium, people developed specialized modes of producing food that led to increased competition for key resources such as pasturelands and for control over productive banana gardens.¹⁰⁵ At this time, people began to refer to unilineal descent groups as vessels for defending rights to land established by firstcomers. Such talk was a new way to think about which persons had the legitimacy and authority to enact ritual control over both ancient and newly established zones of settlement. Propitiating the spirits of a territory controlled by a lineage and its ancestors—the misámbwa spirits—amounted to a claim by homestead heads that a key thread in residential or territorial social identity ran through them. A group of firstcomers who established new settlements at the fringes of older areas of settlement could have used this ritual expression of solidarity and exclusiveness, either at the moment of new settlement or retroactively, to bolster their community as it grew in wealth. This sort of public healing helped those who managed to establish their firstcomer credentials to create, gradually, a group identity tied directly to the lands in which they had settled.

At places such as Munsa, in western Uganda, archaeologists date to the tenth century the earliest phases of settlement surrounding rocky promontories and rock shelters. What perhaps began as a settlement of kin who propitiated ancestral ghosts on the hilltops and in the shelters grew in the fourteenth and fifteenth centuries into much larger settlements of people not necessarily related by kinship, with earthworks that enclosed the favored abodes of the new basámbwa spirits, and sat at nodes of regional trade, including in exotic goods such as glass beads and copper bracelets.¹⁰⁶ The innovations, recognizable in the linguistic evidence, turned on making territorial spirits portable in the zones of older settlement inhabited fifteen hundred years ago by West Nyanza–speaking communities. In building the earthworks, "elite groups both appropriated for themselves and expropriated from their followers the power of the ancestors."¹⁰⁷ The crucial links in this transformation were the medium (the mbándwa) and the institution of kubándwa, through which the ancestral ghosts and the territorial spirits could communicate with their communities of rememberers. West Nyanza societies worked with this supple institution to convert kubándwa mediums of ancestral ghosts into the mediums of territorial spirits.

Following this reconfiguration of mediumship, after the turn of the current millennium, the West Nyanza community fully dissolved into its daughter speech communities of Rutara and North Nyanza. (See Map 2.) And sites such as Munsa grew in size and came to include earthworks. Speakers of Rutaran and North Nyanzan extended the semantic range of the concept of territorial nature spirits tied to a specific firstcomer group by using that term to describe named and mobile spirits responsible for the health of entire genres of experience and environment.¹⁰⁸ As territorial leaders expanded the range of the services they provided to their followers, they drew into their political economies the material sources of wealth that had sustained the smaller-scale firstcomer communities.¹⁰⁹

Map 2. The location of some ancestral speech communities, ca. 1000 to 1500 c.e. See Figure 1 for a partial listing of the languages spoken today in the Great Lakes region that descended from these communities. Map made by Tom O'Connell, Digital Media Services, Northwestern University.

Individual territorial cults could be widespread across the landscape, a circumstance that limited the ability of any single cult to export or project its power to areas where other cults were already in place.¹¹⁰ Territorial cults and the power to allocate institutions of service and tribute transcended the authority of firstcomers in Rutaran and North Nyanzan societies. Wealthy newcomers to a region could curtail the authority of firstcomer groups by creating new nodes of patronage and of military alliance. They used these new networks of followers to redistribute the material wealth in people and livestock that flowed through the increasingly territorially expansive chiefdoms that had emerged during the era of reconfigured mediumship and lineage ideologies just mentioned. Patrilineal marriage alliances created and monitored redistributional networks. Wealthy newcomers to a territory, as a new chief's following, could contest firstcomer claims to its wealth in land and people with new arrays of debt and service relations displayed at events such as weddings and funerals.¹¹¹ Drawing on the close connections between a shrine-keeper's local spirit and its surroundings, these assemblies began to think of the spirits' corporeal forms—watercourses, hills, caves, and prominent rock outcroppings—as sources and categories of their power—fertility, fecundity, war-making, controlling rain, hunting, and so on.¹¹² Instead of ensuring prosperity and fertility in a particular territory, the new territorial leaders might have argued that their new spirits could ensure prosperity in entire categories of the environment and of life.¹¹³

The outcome was the creation of named spirits, such as Irungu or Mukasa, who blended the territorial ritual power expressed in the concept of a musámbwa with the general healing capacities of experts in kubándwa spirit possession. (See Map 3.) In this new context, leaders struggled over extending their power over people (and the land they made productive) and sought to curb the worst impacts of struggle on their own following. The era of ancient state-building had opened with public healing central to its unfolding.¹¹⁴

Map 3. The distribution of some prominent figures associated with the kubándwa form of public healing. The Nyabingi personae appear in a zone straddling the boundary between Ndahura and Wamara (who, with Mukasa, are cwezi figures) and Ryangombe or Kiranga. Reproduced courtesy of Henri Médard, Université Paris I.

Traditions about Mukasa reveal how transforming the territory of healing from local to exportable forms engaged several important aspirations. Mukasa was the spiritual force responsible for healing sickness, giving abundant rain, food, cattle, and children, and ensuring safe passage for fishermen and travelers on Lake Victoria. His shrine was located on the lake's Ssese Islands. Several items from the ethnographic record suggest clearly that his supplicants, priests, and mediums recognized a connection between Mukasa and the musámbwa of Buddu (on the coast opposite the Ssese Islands; see Map 1), a spirit that manifested itself as a python.¹¹⁵ This connection expressed a clear continuity between territorial spirits in general, the misámbwa, and named figures such as Mukasa. And it underscored the central roles of concerns with fertility and the importance of healers in maintaining collective well-being. The power wielded in Mukasa's name by his medium and her priests offered to commoners and nobles alike the possibility of overcoming the death of the body through the birth and rearing of children.

North Nyanzan and Rutaran societies, where these tales circulated, recognized Mukasa as a new sort of musámbwa or mbándwa who was portable and capable of providing all the many sorts of fertility that people required: rain, children, pasture, and healthy soils. They did so around the turn of the first millennium, as the earliest phases of a period of increased humidity settled over the region, perhaps linked to "the Little Climatic Optimum of more temperate regions."¹¹⁶ During this period, major settlements turned up in the formerly dry interior of the region, west of Lake Victoria, indicating a general increase in population densities after the eleventh century.¹¹⁷ And they used a moral and social logic that would have been very familiar to people living as far away as southern Africa.

These changes in portable territorial spirits shaped a third reconfiguration of healing practice in this region. Near the middle of the present millennium, probably during the fifteenth century, as Rutaran and North Nyanzan societies began to dissolve into communities speaking languages that we know today as Ganda, Soga, Nyoro, and Haya, among others, healers and patients working with spirit possession and as mediums invented an entirely new category of spirits whom they called bacwezi. These spirits could return from Ghostland through acts of mediumistic possession, just like bazimu or masámbwa or babándwa. However, the bacwezi did not require their mediums to have a descent relation or a territorial identity with them. These spirits had no lineal heirs; they could assist or possess any individual, in any collectivity, in any territory.

This innovation seems to have been designed to meet a new level of anxiety over fertility. The invention of the term bacwezi occurred after the invention of the term bucweke, which named the condition of an adult dying with no heirs or children of any sort. This word was invented earlier in the millennium, during the end of the life of the West Nyanza community.¹¹⁸ Together with dynastic and clan histories that speak about cwezi spirits—and a long period of pronounced aridity in the region between about 1400 and 1800¹¹⁹—this semantic innovation suggests that during a long fifteenth century, the efficacy of ancestral ghosts, the bazimu, and the stability of lineages was under attack. As agricultural systems struggled to meet demands for food, in the face of growing aridity, people encountered social and political bottlenecks that formed increasingly around hierarchies (some of which were gendered) of access to land, security, and collective well-being. This was the likely context in which cwezi mediums claimed that they could unblock the flow of social life better than the custodians of lineage spirits.

It is important to emphasize the fact that inventing cwezi spirits, building their shrines, and training the personnel who attended to them not only engaged the diminished capacity of lineage spirits to heal, but also enhanced leaders' claims to embody the fecundity and prosperity of their realm, because it offered them a means to extend the reach of their ritual power to a larger community, beyond the language of kinship.¹²⁰ Royal courts sought to co-opt this power of public healing, literally by bringing it inside the royal enclosure.¹²¹ Other royals met this challenge with force, threatening the very existence of independent public healing. These tensions played out in different ways in different parts of the region, but they formed a durable set of poles around which people waged political struggles.

Between the fifteenth and eighteenth centuries, these spirits, and the mediums who could speak for them, came to include departed persons of royal standing, the famous royal Cwezi figures. The historian Renee Tantala has analyzed oral traditions about these new bacwezi and has shown their content to reflect tumultuous contests between the priests and mediums of mbándwa spirits over the importance of ancestral ghosts and the ritual efficacy of kings.¹²² The outcome of these contests shaped the contours of formal dynastic rule in the kingdoms of the central Great Lakes region, surely beginning in the sixteenth century. They led to specialized forms of kubándwa emerging in this period: the cwezi kubándwa in the northern parts of the region, and ryángombe kubándwa to the south. (See Map 3.) By co-opting the ritual practices and priestly hierarchies of these specialized forms of kubándwa, new ruling dynasties hoped to bolster their legitimacy while independent centers of cwezi kubándwa organized resistance to royal authority.¹²³

Late in the nineteenth century, a young Rundi man told the missionary J. M. Van Der Burgt that the word cwezi came from the verb "kucyēra, to spit."¹²⁴ Spitting was a common form of blessing often mentioned in descriptions of practices marking life cycle transitions.¹²⁵ And the Rundi man's derivation tells us that we can think of the cwezi as "the spitters" or "the blessers," as healers who facilitated difficult transitions in life. They were perhaps most helpful to people who struggled with the challenges of infertility.

In the arid period between the sixteenth and eighteenth centuries, these practices must have been extremely attractive to people settling in the dry central savannahs, under the watchful eye of newly emergent monarchies, far from their roots near Lake Victoria and in the Kivu Rift Valley.¹²⁶Cwezi kubándwa offered childless people a setting in which they could develop skills in healing, in part by translating their own struggles with obucweke (childlessness or heirlessness) into succor for others in the same struggle. Although heirless, they were useful people who enjoyed the benefits of being successful healers.

The attractions of this social innovation placed the work and power of mediums (and the priests who managed the shrines where mediums often worked) potentially at odds with the authority of royalty and chiefs. Cwezi kubándwa groups relegated ancestral ghosts to the margins of healing practice and created a structure of initiation that could incorporate large numbers of patients and novitiates. They opened up the politics of prosperity and health in the fissures between courts and their followers. Some royal centers of power appear to have responded by developing a robust military component to emerging state structures.¹²⁷ Where this occurred, it represented the earliest examples of public healing facing threats to its existence.

Figures such as the Nyabingi priests and priestesses appear in tales from Kigezi and northern Rwanda concerned with the growing menace of militarized royal power. Traditions about Nyabingi remark on the fleeting, mobile nature of their power and tell stories about a medium's ability to withstand the most dire force of royalty, and later of colonials.¹²⁸ Steven Feierman has noticed that the image of mediums surviving a firestorm is common in stories of Nyabingi's exploits.¹²⁹ The image appears again, far to the north and east, in Geoffrey Kamali's story of public healing.

Tensions between royal courts and kubándwa groups shaped a fourth era of reconfiguration in Lakes healing practice, stretching from the later eighteenth century into the first decades of the twentieth century. This was the era in which trade and production—increasingly commodified after the 1830s—underwrote and encouraged the militarization of royal power.¹³⁰ Between the 1780s and the 1880s, trade in ivory, slaves, beads, and guns—initially tied to the Nile Valley and the Indian Ocean world, and later tied also to the Inner Congo Basin and the Atlantic—intensified the use of sheer force by kings and others in pursuit of control over this commerce. Between the 1880s and the 1910s, German, British, and African troops undertook the violent colonial conquest of the region and early attempts to establish colonial administrative power. The period closed with the devastating impact on the southern and western Great Lakes region of World War I, and the establishment of different colonial administrations—Belgian and British—in the area.¹³¹

From the 1850s into the first decades of the twentieth century, travelers' accounts and clan traditions speak in detail about the actions of priests, mediums, and the followers of figures such as Mukasa and Nyabingi. They describe women as visible leaders of a critical resistance to royal authority, especially after the turn of the century. At that time, spirit mediums began to turn their attentions to royals who collaborated with colonial conquerors or who otherwise took advantage of the chaos and dislocation created by conquest to prosecute their own aims at political expansion, aggrandizement, and score-settling, and they took on some Christian missions, as well.¹³² But the scenes of public healing composed by priests, mediums, and their followers were not concerned only with the shortcomings of royals.

Whether noted in oral traditions, travelers' accounts, or colonial and mission documents, Nyabingi mediums appear in one or all of three contexts. They appear as healers of individual illnesses, and especially of women suffering from barrenness. In this respect, they appear as extensions of the now familiar kubándwa complex. They also appear as leaders of bands of rebels fighting royal or colonial conquest. And they appear as heroines from the past or as inheritors of a heroic past who have incarnated new power in their present lives. Rastafarian intellectuals drew the Nyabingi figure into their transnational circuits of political activism in the 1930s, not long after British colonial efforts to suppress public healing in East Africa grew in strength.¹³³

As Feierman tells us, oral traditions about Nyabingi mediums often refer to a person with the same name said to have lived in different times and different places. The example of "Gahu" illustrates the temporal and spatial mobility of single identities. Some traditions say that Gahu lived as an early-eighteenth-century queen in Mpororo (in southern Uganda); others say she died in 1931, in northern Rwanda; and still other accounts, collected in the 1970s, say that she was still alive then, living near Byumba in northern Rwanda.¹³⁴ These traditions refer to mediums such as Gahu dying, returning to life, and dying again only to make possible still another return.¹³⁵ Feierman and the Rastafarians, more than any other historians of Nyabingi, have teased an important conclusion from these tales about Nyabingi's priestesses. Feierman finds that the tales act as charters for the "instability of authority" and reinforce an old argument in African politics, "that chiefs, lineage elders, and kings hold power because they've seized it."¹³⁶ Nyabingi's power mirrored chiefly and royal power, but it did so in an unstable manner.

Nyabingi's female figures appeal to scholars because they shaped fields of political struggle parallel to and in conflict with royalty, and which colonialists and Christians tried to shape or even to destroy.¹³⁷ Yet the visibility in the colonial archive of public healing such as Nyabingi's cannot be explained solely in terms of conquest and proselytizing. Historians have found that the disastrous epidemics of the 1880s and 1890s and the dramatic rise of sexually transmitted diseases in the 1900s renewed the need for public healing.¹³⁸ The slave raiding and caravan trading of the middle of the century dislocated people, changed the prospects of young men and elite women for engaging in commercial activity and social climbing, and radically altered the capacity of royals to maintain control of the commercial networks that helped underwrite their authority.¹³⁹ Epidemics of rinderpest and sleeping sickness—devastating to livestock and people—broadly accompanied colonial conquest and continued to shape Africans' demographic health and food security long after colonial armies had become colonial police forces. These processes of profound social and ecological rupture framed the actions of Nyabingi mediums and their networks of patients against royal and colonial military force.¹⁴⁰

Acts of public healing directed at individuals, at military power, and at entire collectivities of patients and other mediums elicited efforts by colonial and mission agents to co-opt, suppress, or destroy them.¹⁴¹ By early in the twentieth century, "authority over public efforts to bring health to the populace shifted from African to European hands."¹⁴² But Nyabingi's people are not dead. Rumors of their power circulated among colonials in the 1930s, and mediums continue today to direct public healing events in East-Central Africa.¹⁴³

Consider the efforts of Jjajjà's group of supplicants from the perspective of a long-term regional history of public healing, and their social and moral logics blend with historical narratives of capitalism without ceasing to be "fit explanation for contemporary situations."¹⁴⁴ In Kamali's scene, people pursue largely individual aims and aspirations, such as personal monetary gain and individual bodily health. The particular forms of individualism that emerged during colonialism are in play here, but their fate is tied up, in part, with the practices of public healing. The political collectivities of Nyabingi's followers, or, in a still earlier era, of a mucwezi's followers, echo in the crowd of supplicants and gawkers—a moral community of seekers with a public face. They realize their individual aims through a collective action that promises moral continuity and material consequence in the face of "a world gone awry."¹⁴⁵

Larger changes in economy, society, and polity shape durable motive and action in contemporary public healing. But these changes are not things that have already taken place in Europe, or in the wake of Europe's history, and which here merely take a local form.¹⁴⁶ Christianity, colonial conquest, waged labor, and electoral violence ushered in modernist African aspirations that appear at once moral and material, spiritual and practical.¹⁴⁷ The motives for engaging a colonial modernism simultaneously lie in long-term regional histories where religion and politics were not distinguished as such until the nineteenth century, and in more recent entanglements with individualized dimensions of producing moral persons. Neither the market's autonomous individual nor the anxious jealousy of the accusation of sorcery can explain fully "the eminently public enchantment of legitimate authority and its capacity to crystallize and channel the energies of moral collectivity."¹⁴⁸ This claim may seem unremarkable, but only because it is often established for contexts in which important themes of historical change—such as Protestantism or capitalism—with origins in Europe are in play. In practice, histories of what motivates the pursuit of health and healing in Africa are very rarely approachable from a position informed by a rich history of deep chronology and great regional extent.

African regional histories offer a means to understand their experiences and assess future actions in a world of flows of people, culture, and capital using discourses and practices whose great antiquity reflects their sustained capacity to articulate and deliver what people want.¹⁴⁹ If forms of reason and practice with roots in the African past have effects upon the present moment, we need not see them as relics from a past whose time lies before the modern. But does their existence disrupt a notion of the present as a holistic time—the culmination of "the useful but empty and homogeneous chronology of historicism" that somehow renders the present free from the past?¹⁵⁰ The historical development of durable forms of power and action across ruptures in the past reconfigures "the conceit that the world expansion of capitalism brings all other cultural history to an end"¹⁵¹ by juxtaposing the expansions of capitalism with theories of power, of objects, and of the social, set in temporal and spatial frames grounded in Africa.¹⁵² Such African regional histories refuse a tendency to totalize the transformations that mark historical periods, even while the new history takes shape inside a particular set of master narrative conventions.¹⁵³ Precolonial African regional history helps portray Africa today as both "modern" and otherwise, but not as the result of some a priori difference between Africa and elsewhere.

Traditionalism studies cannot replace modernity studies, if only because the two terms sustain each other. But if scholars unpack select notions of the traditional, implicitly labeled as such by claims to modernity, they open up a curious new perception of the present. Historicizing the social configurations and intellectual underpinnings of, say, "African states" as deeply and broadly as sources will allow reveals a dens