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: 22 September 2023

Early African Pasts: Sources, Interpretations, and Meaningsfree

Early African Pasts: Sources, Interpretations, and Meaningsfree

  • David SchoenbrunDavid SchoenbrunDepartment of History and Program of African Studies, Northwestern University


Writing Africa’s history before the 10th century almost always means relying on sources other than written documents, which increase in number especially from the 16th century onward. Archaeology (including the study of art objects), the comparative study of historically related languages, paleo-environmental studies, and oral traditions provide the bulk of information. Writing Africa’s early history ideally involves collaboration among experts in using each kind of source, an increasingly common practice. Despite the challenges of analysis and interpretation posed by this base of sources, early African history has a depth and breadth akin to the histories made from the written sources in archives. Even so, whereas written documents provide details about individuals and precise dates, the sources for writing early African histories more often provide detail about conceptualization, for example, of time, hospitality, and individualism and about larger, environmental contexts shaping those concepts and shaped by the actions of the people who held them. Translating such concepts and scales of action into accounts accessible to those—including many historians—not steeped in the methodological conventions underlying the analysis of each source is a major challenge facing historians of Africa’s earlier past.


  • Archaeology
  • Central Africa
  • Cultural History
  • East Africa and Indian Ocean
  • Historical Linguistics
  • Historiography and Methods

To write about periods and places beyond literacy, that is, without fine-grained calendar dates and named individuals, historians developed a conventional base of sources. Paleoecology, historical linguistics, archaeology, and anthropological genetics help most with the periods before the 10th century.1 Oral traditions, a great variety of (representational and functional) objects, ethnographic descriptions, and other documents help most with the following centuries, the latter two sources being more numerous after 1500 at Africa’s oceanic and desert edges.2 Each source reflects the past in different ways, shaping the historian’s interpretation. Each source is dated with varying degrees of precision adding further challenges to the historian’s task of exploring and explaining change and continuity over time. The sweeping scope of the resulting narratives can feel cold, for it resists a reader’s projection of self into them. Nonetheless, such narratives focus on people’s aspirations and the forces limiting their pursuit of them. They track the incremental accumulation of the consequences, intended or not, of pursuing aspiration in particular ways. And they capture the ruptures that can create new awareness of change and continuity for actors, reshaping what people understand as aspiration. These conventions of writing Africa’s earlier history should be familiar to scholars steeped in the written word and surrounded by things and places easily pinned down in a fine-grained chronology. This article explores what historians can say about the past, using each source, in order to make their narratives of Africa’s earlier histories more accessible. That is a precondition for taking Africa’s distant past seriously. The essay focuses on Africa east and south of the Inner Congo Basin (see Maps 1 and 2) because that is the richest part of the continent in the mix of sources and narrative just mentioned.

Intimate Climate Change

Ancient ecological studies detail the interplay of climate and vegetation change, two important variables in early African history. Approaches to that complex dance of influence and change have shifted in the last decades. For example, ecologists used to think about change over time in plant communities as something that moved through an orderly succession of steady-state communities, leading to a climax community of species, a stable form disturbed only by external factors—such as human activity or climate change. This view has given way to a richer sense of the interactive forces lending a hard-to-predict dynamism—rather than a natural tendency toward stability—to changes in plant communities. In this view, many life forms, people among them, produce change more or less constantly. The disappearance of the older notions opened the way for ecologically inclined social scientists to recognize, for example, that forests—or parklands—might be the product of people curating particular plant communities rather than a steady-state of particular biological processes.3 Much of that kind of work in African history has focused on very recent times, but the approach requires thinking about environmental and social change together in the earlier African past.4 As the idea of an anthropocene—the period that began roughly at the turn of the 18th century, when human industrial activity become the principal driver of environmental change around the world—gains ground, further research into earlier, regional iterations should expand.5

For example, periods of reduced rainfall can magnify the effects of farmers and metallurgists clearing forests for field and fuel. Longer rainy seasons can raise the turbidity of a lake’s water stack through shifts in prevailing winds, or of a river’s braided course through prolonged flooding. Shifts in the onset of rainy seasons, and fluctuations in the volume of rainfall and the length of the season prompt new emphases in the mix of crops, domestic animals, and the importance of fishing, trapping, hunting, and foraging in regional food systems. For the lands surrounding East Africa’s Inland Sea (Lake Victoria; see Map 1), many such studies exist, each set in one of the region’s different environments—montane, upland, forested, swamp, or lake. They yield a composite picture of large-scale patterns and regional variations in the nature and timing of these shifts. Many can be given calibrated radiocarbon calendar dates, lending them a welcome degree of chronological precision. At tighter scales of time and region, shifts unfold quickly, intimately, the sort of shifts that people recognize.

Map 1. Africa: Physical Features and Current Nations Mentioned in Text.

One major event—a change in rainfall seasonality—centered in the 12th century, stands out in this regard. West of the Inland Sea, a brief period of increased moisture had opened in the 10th century before a long period of aridity set in, from 1040 to 1240, around the region’s crater lakes.6 The richer moisture budgets in the 10th century, followed by the dramatic drought of the 11th and 12th century that affected all of East Africa, coincided with the onset of the so-called “Medieval Warm Period” in Europe.7 The equatorial expression of another, later large-scale climatic process, the onset of the “Little Ice Age” generally associated with “improved moisture balance,” began in the 13th century, with two East Africa phases.8 Compared to the previous several centuries, these changes left the region with a pronounced seasonality to its rainfall regimes.9 The onset and volume of the short rains became uncertain. The water budgets in the crater lakes and swamps west of the Inland Sea shrank dramatically for about four human generations, from 1050 to 1150. The Inland Sea’s water budget shrank about a century later, in 1140 ce. Between 1140 and 1160, the first big drop in the water’s level was recorded. A brief uptick followed, between 1160 and 1180. For two decades, from 1180 to 1200, littoral communities witnessed the biggest single drop in water level of all of those that occurred between 1040 and 1360.10 During those two generations, from 1160 to 1200, climatic familiarity disappeared altogether, just as it has for many of us today.

Climate and vegetation change was intimate for early Africans, but some of it affected large areas in a similar fashion. From the Northwest corner of Africa’s Equatorial Rainforests to the Northern Kalahari’s Okavango Delta, between 4,000 and 2,500 years ago, a turn to aridity that peaked with the onset of a more pronounced dry season joined the persistent intercalary semi-deciduous rainforests—common on patches of the ancient Kalahari’s sandy soils—into a corridor free of a closed canopy of tall trees.11 Changes in this so-called “Forest-Savanna Mosaic” are documented in pollen cores, taken from lakes and bogs across the region, which contain different layers and volumes of the pollen from trees and shrubs common in distinct plant communities.12 By sampling the layers and dating those with enough carbon in them (from burning events) to yield radiocarbon dates, one produces the sequence of ratios and rates of vegetation change glossed here as a gradual joining of savanna patches into a continuous piece crossing the rainforests. Large-scale climatic shifts—not the actions of people—drove this process. But, this was the same time and place in which early Africans transformed the continent’s linguistic geography on a scale of impact as profound as—but even broader than—the spread of Romance and Germanic languages had on Celtic.

Words, Things, and Meanings

This story has been tethered to a glaring fact about Africa’s linguistic geography. Of Africa’s more than 2,000 languages, the 550 or so spoken south of the equator, share a larger amount of words and grammar than any other similarly sized group. That mix of similarity and geographical extent has prompted historical exploration since the 19th century, recently reaching consensus about the routes of spread accounting for the current distribution of all those languages, called Bantu (see Map 2).13 The close of the 19th century was also the opening of Africa’s short colonial century, so the topic that has come to be known as “the Bantu expansions” has had to shake free not only from forests but also from a past in which linguistics played a powerful role in constituting colonial and national categories.

Map 2. Bantu Zone.

Linguists increasingly took note of Africa’s languages in the 19th century. In their fraught efforts to evangelize and rule, missionaries and colonials needed knowledge of African languages. They applied the historical sensibilities of their time to the realities of a continent engaged in commodification, evangelism, the bankruptcies of slavery, and the dislocations and destructions of imperial conquest.14 These contexts shaped a rich archive of texts by outsiders, of vernacular texts, including ethnographic accounts by the author of their own cultural practices, sometimes in an historical voice, and dictionaries that inform the conventional reconstructions of words and meanings, and speech communities—including Bantu—that provide such important evidence for Africa’s earlier history.

Establishing older speech communities begins with collecting and comparing vocabularies from each language in a historically related set, like the Romance languages. Scholars apply the comparative method to these data to discover patterns that guide inferences about sequences of change and continuity in the forms and meanings of words.15 The method compares vocabularies from languages suspected of sharing historical ties to infer earlier forms of their pronunciation and meanings. The researcher usually generates words through elicitation work, or draws them from dictionaries and vernacular texts.16 The goal is to look for regularly corresponding sounds in words from different languages that have the same or very similar meanings. Those patterns reveal cognates and help distinguish them from innovations, allowing the tabulation of each.17

Tabulation facilitates sorting the languages into smaller groups, based on the cognates they share, but confirmed by lexical or grammatical innovations shared only by members of each group.18 One idea is that languages sharing higher rates of cognation have had less time pass since they began to diverge. Languages with lower rates of cognation have had more time pass since they began to diverge. More elapsed time means more opportunities for new words to replace older ones. The other idea is that, if a set of languages shares unique grammatical or phonological features and/or vocabulary, the simplest way to explain that is to propose their common origins during the life of the speech community from which the given languages descended. Innovated material confirms (or calls into question, if no such material may be adduced) the historical existence of a common speech community first hinted at by patterns of cognation. The analysis requires having exhaustive distributions of the words, sounds, or grammatical features in question and on identifying the patterns of sound correspondences that sustain a claim of cognation and allow the recognition of innovation. Africa’s linguists continue to work to fill the many remaining gaps in this kind of high-quality evidence from the continent’s languages. The sorting process produces a sequence of language or speech community divergences, beginning in the past and ending in the distribution of languages the researcher encounters in the present. A speech community is joined the minute one speaks a member language. It must not be confused with—or reduced to—the socio-political dynamics of inclusion in and exclusion from an ethnic group, even though language is often a strong marker of or medium for ethnic process, especially through literature. It is the historian’s task to explore the ways in which language and ethnic formation mutually inform each other; we must not assume that influence. Identifying and counting cognates, sorting languages into groups, then identifying innovations shared only by members of each group that confirm (or revise) those groups, and sequencing their formation and dissolution constitute the method called “Words, Things, and Meaning” (hereafter, WTM).

This method relies on three kinds of imprecise chronology. One assigns a temporal value to the rate of random accumulation of word replacements in a list of core vocabulary, the kinds of words least likely to be borrowed. The randomness of word replacement is indicated if plotting cognation rates between each two of the total set on a graph yields a bell curve. The center of that range reflects the amount of such random accumulation of lexical replacement connected to that particular group of languages. This is called glottochronology if one converts that range into calendar dates for the language divergences represented in historical classifications.19 The second option accepts the relative chronology implicit in sequencing the language splits. It provides no calendar dates for the divergences. Last, one can reconstruct words to specific intermediate speech communities, signifying things like an iron tool, a particular domesticated animal, or a species of fish, for which we have radiocarbon dated archaeological evidence from sites within the territory arguably occupied by people in the given speech community.20 In this circumstance, the range of calendar years generated from archaeological evidence also dates the existence of the speech community whose vocabulary contained the reconstructed word and meaning.21

Early formulators of glottochronology derived a constant rate of lexical replacement. But, languages are always changing and the rates of change reflected in bell curves vary. Critics of glottochronology find such variation refutes the notion of a constant rate of change, thereby rendering the technique useless. But, if variation in the median of bell curves can be calibrated by direct association, scholars can raise questions about the social contexts and the arrays of standing and creativity that affected different rates of change in different parts of living language. The recent arrival of consensus in the classification of Bantu means scholars can focus on direct association to generate a subtler approach to rates of change, nearer to those with which lexicographers and sociolinguists work.

Scholars classified Africa’s languages first to establish the internal historical relationships reflected in diverging speech communities, their histories of interaction reflected in lexical transfers, and the reconstructed vocabulary each ancient speech community used. The goal was to generate a basic sense of where and when people speaking these languages lived in the past and to present a basic sense of the social, economic, cultural, and political lives they lived. Such histories of settlement and basic cultural practice, inferable from classifications and lexical reconstruction, could then be set alongside what archaeologists had learned about the same topics and regions.

The idea is simple: if a verb for “to smelt iron” may be reconstructed to a particular ancestral speech community, then people in that community knew about iron smelting. The existence of the word in the past implied the existence of the thing to which it referred and, so the logic goes, the archaeological record should confirm or deny something about that knowledge, beyond dating it: was the knowledge direct and functional—did the people speaking the language with the verb “to smelt iron” actually smelt iron—or were they simply familiar with iron smelting from anecdotal sources? Where the two records of the past are equally rich, such as in east Africa’s Great Lakes region, its Indian Ocean coast, or in southern Africa from the Zambezi basin to the Cape of Good Hope, comparing them offered the richest rewards to historians. However, in central Ghana, or the Inland Niger Delta, or the stone hills of the Bandiagara escarpment, on the Delta’s southeast, not only does the archaeological record dominate, but the linguistic map is far more diverse than it is in much of the Bantu-speaking parts of the continent. This situation requires collaboration with experts in each of the different families of languages represented.22 But, such collaborations rarely happen; the value of comparing language history and archaeology lies mostly in framing paths for future research.23

Recent collaborative studies of the expansions of Africa’s Bantu languages have used aspects of WTM to reach consensus about the broad geographical outlines, if not the timing of and routes taken in that vast story.24 For example, knowledge of particular species of trees, indicated by firm lexical-semantic reconstructions, correlates with paleo-environmental evidence, indicating shifts in the distribution of the varieties of forest in which such trees naturally occur.25 In the northwestern zone of the vast region in which Bantu languages are spoken, each of three species of tree has two different names. But, for each pair of words, one distribution crosses the other, implying a relative chronology and a circumstance for renaming. For example, the species of tree botanists call Musanga cecropioides was well known as “*-gùmbù” to the ancestral speech community that was bisected by a younger speech community, which renamed the tree “*-céngà.”26 The tree grows in regenerating forests between savannah and closed-canopy forests. Thus, in the initial phases of expansion, Bantu-speakers avoided closed canopy rainforest environments favoring older patches of savannah or semi-deciduous open forests and the emerging corridors joining them.27 The archaeological presence of Pearl millet (Pennisetum glaucum) in areas today dominated by rainforest, and the ability to reconstruct a root, *-cángú, with the same meaning to the most recent common ancestor of subgroups whose ancestors lived in these ancient savannah corridors strengthens this account of expansion.28 The social dynamics of Bantu-speakers gaining knowledge of life in forested environments from earlier groups influenced speech communities throughout the larger story, initially in savannah and semi-deciduous forested environments. These braided threads of evidence settle on a single model for the expansions of Bantu languages (see Map 2), in which the earliest expansions unfolded in the northwestern regions over a much longer period than the subsequent expansions, which unfolded south and east of the Inner Congo Basin’s thick, shifting forests.29 But they paint the social complexity entailed by that long history in strokes so broad that ecological processes—not people—play the role of contingency historians value.

At smaller spatial scales and shorter spans of time, that perception fades.30 Working at smaller scales, linguistically inclined historians have pushed beyond establishing settlement sequences and chronologies and basic outlines of economic practice to explore social relations, political culture, and technology. As a result, they have placed much more interpretive weight on the meanings of words in the past than on the implications of classification for sketching a settlement chronology. For example, all four languages constituting the North Nyanza subgroup, spoken on the Inland Sea’s Uganda littoral (see Map 3), have a word, *kika, that can be glossed in English as “clan.” The regularity of the term’s pronunciation, its distribution, and the consistency of its referring to the dispersed communities that can constitute a clan implies the word and meaning existed when Proto-North Nyanza was a single intercommunicating speech community, between the 8th and the 12th centuries.31 All the (19th and 20th century) descriptions of social settings, in which the category called *kika is in play, also reveal that membership was marked by shared totemic avoidances, unique drum rhythms and children’s names, as well as claims to descend from a distant, shared ancestral figure or figures of either gender, though most are male.32 In languages with this word, people pronounce it using common sequences of sounds used in pronouncing other vocabulary. But the tonal qualities of pronunciation differ regularly between the languages. Thus *kika was part of the lexicon of Proto North Nyanza, out of which the languages today called Luganda, Lusoga, Lugwere, and Rushana diverged.33 But it was a new part, and the role of contingency in that novelty is reflected in the semantic resources people used to make the new word.

Map 3. The Northern Inland Sea Islands, North Nyanza, and Rutara.

Calling clan networks ekika, a thousand years ago, was an innovation that can be dated and, thus, given a broader historical context. It was accomplished by applying the singular prefix (ki-) to the noun stem (-ka). When –ka, the noun stem, occurs with the plural prefix (ma-) its referent includes “homestead” or “family.” This happens in languages beyond and within the North Nyanza languages. The use of a particular singular prefix (ki-) to mark the meaning “clan, patriclan” was, in effect, a proposal that such dispersed groups were like a big family residing at one homestead. Today, this is no longer a “proposal,” it has settled into ordinary speech in North Nyanza languages and a few immediately adjacent languages. But, the distribution allows us to infer that the proposal was first offered sometime during the period in which Proto-North Nyanza existed as a speech community, because the new term and meaning that resulted—ekika (clan, patriclan)—is limited to North Nyanza languages, while the term amaka (homestead or family) is found in North Nyanza languages and in other, non-contiguous languages descended from Great Lakes Bantu. Both terms and their referents have distributions conforming to subgroups set up using the comparative method; the first is narrower (and younger); the second is wider (and older).34 The context prompting the invention of a new kind of clanship had to do with fostering a new sort of organization, hinted at in the semantics of the innovation. A century or so on either side of 1000 ce, on the Ugandan littoral of the Inland Sea, it began to make sense to people to think of their clans as families, rather than as undifferentiated groups. This welcome shift of analysis, from the unforgiving qualities of climate and vegetation change to the taxing vagaries of social life, still begs the question of the fates of the individuals in these groups. South-central Africa’s on-going Bantu expansions, of the same time period, offer a rich place to do just that.

The figure of the individual and the multi-lingual regions individuals socially composed both come into view through WTM. The historian Kate de Luna classified the languages spoken today from Okavango to the Batoka Plateau (see Map 1) and reconstructed significant portions of their vocabulary, in the manner described above. In de Luna’s history, famed spear-hunters, spear-fishers, and metallurgists, who worked away from settlements, disturb the inertia of agricultural processes driving accounts of Bantu expansions.35 She tracks the innovation after the 8th century ce, in speech communities descended from Proto-Botatwe, of a new word and meaning “bush.” This new conceptualization of the landscape separated domesticated fields and settlements from the bush, where valuable materials were won. Other innovations, made in the same speech communities, named kinds of hunters and metallurgists famed for their skill and generosity. In the eyes of ordinary fishers and farmers, these were the individualists of the 8th to 13th centuries ce, working the danger zones of the middle Zambezi and Kafue river valleys (see Map 4). They were drawn to the emotional charge of recognition earned by success, both in their lives and as ancestors whose accomplishments warranted the memory work of those who came after them.36 Audiences likewise found in the spectacle of abundance, bravery, and knowledge, a reassuring confidence in their compositional strategies of accumulating expertise and diverse networks of skill.

Map 4. The central Indian Ocean Coast from Lamu to Mozambique

Mobility and worldliness drove that practice, a conclusion supported by the fact that key terms in a regional lexicon of bushcraft were borrowed from—or innovated together with—Bantu-speaking communities neighboring the Botatwe zone.37 For example, between the 600s and 1200s ce, their industry in metals (especially copper work), ivory, and smoked water antelope meat—and their interests in mobility—drew the specularite (a shiny mineral crushed and used as a cosmetic) and glass beads of the Western Kalahari and Indian Ocean worlds, respectively, into theirs. Movement required the exchange of goods, knowledge, and social ties, enriching the meaning of trade and revealing that Botatwe-speakers sought status through a “political culture valuing worldliness, fame, wealth, and skill.”38 Distributions of words for skilled hunter, fisher, or smith cross the boundaries of Botatwe speech communities, reflecting all the connections. De Luna’s work reveals the lively creativity of the inheritors of the earlier expansions of Bantu-speaking communities. The contingencies of social facts like the gender of bushcraft, the affective charge of seeking status, and practices of belonging all work with the stabilities of ecological processes. But, they revise the study of systems, such as kinship or social complexity,39 leaving us with a pleasing sense of individualism and worldliness at the core of possibility and constraint in the distant African past.

People speaking descendant subgroups of Botatwe, east of the middle Zambezi, developed this social life at the turn of the first millennium. It facilitated a 360-degree orientation to shifting centers of wealth and power on the sometimes distant, sometimes overbearingly close edges of their central frontier.40 Noting which of the several centers of artisanal and political skill shares with Central Eastern or Kafue Botatwe-speakers particular bits of the language of bushcraft and fame reveals bushcraft’s eclectic nature. They shared with Njila-speakers, to their west, a fashion for hunting associations. They shared with Sabi- and Luba-speakers, to their north, a respect for the talent and skill of the famed, especially hunters who worked in groups with iron-tipped spears.41 They shared with Kusi-speakers, to the south, a set of terms for the natural environment, new to Proto-Kafue speakers.

Exchange networks on the central frontier, in the 16th and 17th centuries, looked to both oceans. But Botatwe-speakers lived as far from the oceans as anyone, free from direct control by coastal centers, old and new. That freedom is eloquently attested in the abundant “foreign words, objects, and styles” they “used to assert, display, and aspire to another social status.”42 So, WTM brings to light the single speech community—whether the ancient “Proto Bantu” or one of its many, younger descendants, like “Proto Botatwe” or “Proto North Nyanza”—as a locus of narrative. It also reveals zones of cultural bricolage—learning from a variety of cultural traditions and blending what you like from each—shared by more than one speech community.

Mobility and bricolage also drove historical change around the rim of the Inland Sea. The distribution of the term musambwa crosses the boundaries of linguistic subgroups, and the prefixes attached to the stem differ. But in each case, the word refers to a kind of spirit whose abode and efficacy is territorially limited. In each case, the noun may be derived from the passive construction of a verb *-samba, “to kick,” making plain that a person possessed by such a spirit had had an experience akin to being kicked.43 These features suggest very strongly that the word and its referents were spread around the Inland Sea through social intercourse. If the term had been part of the Proto-Great Lakes Bantu vocabulary, and if it was inherited in each of the subgroups that settled around the littoral as the Great Lake Bantu speech community dissolved, then the prefix should be consistent, as it is in the case of amaka or ekika, discussed above. It is not consistent and that suggests that different local speech communities on the littoral took up the term separately, with speakers in each setting using different prefixes to integrate the idea and practice into their vocabularies. The important point for historians is that, before new words and meanings become literal meanings—for [bush] or [famous person] or [clan] or [spirit]—for their users, they are improvisational responses to particular problems or settings across a wide region.

Focusing on change and continuity in meaning, linguistically inclined historians have uncovered dimensions of value and logics of social practice. Rendering speakers of past languages actors in their own stories personalizes the early African past, bringing it closer to today’s readers. Arguing lexical and semantic reconstruction implies a durability of communicative context over time. But, the abstractions beneath reconstructed words limit what historians may ask of recovered meanings. Many reconstructions are exceedingly vague, pointing to little of the social contexts of their use while others, like musambwa (possession by a spirit), point directly to contexts of use.44 Reconstructing words with fields of meaning that overlap and mutually inform—such as eka (homestead) and ekika (clan)—and locating them in an ancestral speech community’s vocabulary also helps recover contexts of use. But, the formalism of WTM struggles against the paucity of evidence concerning the relational core of communication—conversation and reflection—in which meaning lives. This struggle for context where evidence is scarce, and a strategy for mitigating the scarcity, is taken up in a little more depth in the last section. Getting at past contexts of using language, objects, and space helps avoid the “view-from-nowhere” that systems like language can smuggle into analysis and interpretation. One way for historians to mix into a system’s rules the surprises of practice that give contingency life45 is to learn from archaeologists’ attention to variability in material and spatial patterns.

A Socialized Archaeology

Archaeologists, along with ecologists and historical linguists, have come to appreciate the ambiguities and contingencies of accounting for change and continuity in the human past. They once reconstructed whole cultures, using the material and spatial patterns revealed in excavation and survey as proxies for concepts such as [state], [chiefdom], [farmer], or [forager] or for ethnic groups. These categories were invariably understood by reference to written accounts of and/or research on societies thought to fit these descriptions. The anthropology that supported this rather presumptuous approach to cultural pasts grew much more sensitive to dynamics of change and conflict. Not later than the 1950s, as many of the colonial settings in which anthropologists had worked shook loose the bonds of foreign rule, archaeologists increasingly recognized that material culture does not necessarily signify unchanging features of an ethnic group, a particular political form or economic orientation. The challenges of documenting change and continuity should focus instead on tracking the dynamic and contingent interplay of environment, economy, intellectual and political culture.46

Even this important advance could promote a sense of linear change in the scale and complexity of social life, falling prey to thinking in terms of stages in such development, from small to large, from few people in a society to many thousands.47 Or it could fall prey to the idea that the meaning of the things and spatial relations that excavators unearthed and surveyors mapped could be clearly understood in a straightforward juxtaposition to more recent, documented settings. This last risk always attends a method, which specialists call the “direct historical approach,” of using so-called “ethnography”—accounts of culture (often but not exclusively authored by strangers to that culture) with some arguable historical connection to excavated and surveyed materials.48 Simply demonstrating such connections does not go far enough in exploring and explaining the differences between them. Failing that, one risks anachronism—projecting into the past the whole cloth of what has been documented ethnographically, only much more recently.49

To protect against these tendencies, archaeologists embraced uncertainty in the attribution of meaning in the distant past, set aside notions of holistic social progress, and took up meaning-as-action—consuming, making, exchanging, and so forth—and social practice as inflected by social statuses like gender, age, and so forth. They take very seriously the difficulties of knowing about meaning in early history when context and speech—in a word, communication practices—eludes us.50 The results have brought archaeological research and writing close to historians committed to depicting the particularity of cultural experience and the exceptionality of sequences of change and continuity before seeking to generalize. Africa’s archaeologists have been at the forefront of each of these intellectual movements.51

The rims of the Inland Sea and of the Indian Ocean, like de Luna’s central frontier, have rich archaeological records of movement and the exchange of ideas and things amenable to these new methods of interpretation. On the Indian Ocean coast of East Africa (see Map 5), whitewashed mosques, tombs, palaces, great houses of coral and lime mortar, and accounts of proper Islamic piety and hospitality penned by august visitors have long attracted the scholars keen on cosmopolitanism and early forms of global history.52 Recent work has foregrounded the social effects of architecture and objects as media for experience, not only as advertisements to others of the social reach of networks.53 The maritime orientation for which the region is justly famous took centuries to manifest fully. Stone-built urban form addressed to the sea emerged after c. 1000 ce.54 An abundant vocabulary of ocean-going life forms—sea turtle, shark, whale—is restricted to Swahili, which emerged as a distinct language by the latter centuries of the first millennium ce, with numerous dialects emerging thereafter along the coast, from Southern Somalia to Northern Mozambique.55 Between the 11th and 16th centuries, Kilwa Kisiwani (Map 5, inset b) was the premier coastal African entrepôt that connected, among many others, Botatwe-speaking hunters and smiths with the glass and shell beads of the Indian Ocean world. By then, the sea was as important a historical horizon for East Africa’s coastal people as was their hinterland.

Map 5. The greater Botatwe zone, Northern Kalahari, to central Zimbabwe.

Confining their material productions inside a single ancestral speech community is easier said than done. The earliest archaeological iterations of the coastal cultural world popularly called “Swahili” are found in central Tanzania.56 But, northern Kenya is where the diversity among Swahili languages is greatest, implying that their ancestral speech community lived there.57 Central Tanzania’s littoral would have been the eastern edge of the territory in which people spoke Proto Northeast Coastal Bantu, the speech community out of which Swahili eventually emerged.58 The formation and divergence of Swahili, then, happened later, with the benefit of a few centuries of learning about the Indian Ocean. Whether coastal history was driven by mainland or maritime connections provides the familiar drama and complexities of a borderlands history drawn at the edges of land and sea. But, the importance of the built environment, the objects it housed, and their users’ language, make the drama intimate.

It was a drama oriented to visitors. Along the coast people built towns and villages from earth and wood for many centuries after they first turned up there. Don’t be fooled by such unassuming building materials. Inhabitants did not limit their sights to local settings. From the moment the towns and villages appeared, early in the 500s ce, they imported things like glazed jars from overseas. On Pemba Island, imports and local craft production are distributed throughout the earliest sites, suggesting very broad participation in the market for commodities.59 But, at Shanga and Kilwa Kisiwani, far to the north and south, imports are rarely found far from a stone-built center. All such towns were engaged with overseas trade, but not in the same ways and not with the same list of trading partners.

The glass and porcelain imports bear chemical signatures unique to workshops in the Mediterranean, the Persian Gulf, South Asia, and China, where artisans from or influenced by India and others worked.60 They are famous for demonstrating East Africans’ involvement in the affairs of those distant places. But they form less than 1percent of the total volume of objects archaeologists have excavated from the towns and their hinterlands. Their value, then, was clearly as rhetorical—invaluably so—in the distant past as it is now and makes sense only against the common world of local pottery, cotton cloth, iron, and ground shell beads. At the end of the first millennium ce, that common world of things began to shift. Glass beads displaced locally ground shell beads, Kilwa began to mint coins, with the copper issue circulating as far as Great Zimbabwe (Map 4), and bowls displaced jars as the preferred way to use imported ceramics: for eating a proper feast of rice and fish or millet and beef.

At the same time, some people took up building in coral and lime, a fashion first used in 10th-century pillar tombs, which had spread by the 11th century to mosques, other large public buildings, and elite town homes.61 Many older earth and thatch villages took no interest in the new building fashion, but at places like Kilwa Kisiwani, Songo Mnara, or Chwaka, stone buildings—and the spaces they defined—shaped urban life. The use of bowls to feast near a mosque and the desire for beads of glass from India or the Persian Gulf, rather than beads of local shell, raises questions about the entanglement of affective and intellectual life with public eating and the importance of worldliness as a symbol of standing and authority.62 Imports, properly displayed and used with locally produced things like ironwork, shell beads, pots, ivory, copper jewelry and boxes, and possibly gold, bundled the enormous range of a high-status feast-thrower’s network with the sensorium of the tasty food and drink to make one concise statement about the importance to standing of extravagant generosity. The decorated, locally produced single-serving bowls tumbled together with some large serving dishes, found by Stephanie Wynne-Jones at Vumba Kuu, near the one stone-built mosque in that 15th century town,63 testify more elegantly than a traveler’s memory of hospitality to the fraught fun and fame of a meal catered for many by a leading figure in the town’s urban life. It contrasts with embedding Chinese porcelain bowls in the archways and domes of mosques and palaces, as at Kilwa Kisiwani or Songo Mnara, or using the bowls to serve food to guests in the courtyards of the great houses in those towns. In each place, rich or not so rich, largesse and hospitality went hand in hand.

It is common to assume that, as coastal people embraced Islam, their broader cultural orientations—not just religious ones—conformed increasingly to other parts of the Islamic world. But the Islamic world and the Swahili world were not uniform then, as they are not now. Practices of gendered seclusion, common from the 18th century, as Omani interests found homes on Lamu and Zanzibar islands, are not in evidence at Songo Mnara, near Kilwa.64 Sightlines connected the rooms in Songo Mnara’s 14th-century great houses.65 Spinning cotton thread and grinding beads occurred in all rooms.66 Traders back from overseas—or visitors just arrived—were welcomed into the intimate spaces of their hosts’ houses and entertained with a beverage in a stepped court whose niched walls swarmed with imported bowls, jars, and, perhaps, a Qur’an, proclaiming the standing of the hosts. Eating of certain foods, in certain bowls, at certain times of the year, hosted by certain people of standing—or by newcomers aspiring to do the same—reveal regional contours of similarity and variation.

The archaeological record for the Inland Sea’s littoral (including the lands in which Proto North Nyanza was spoken) contains evidence of change and continuity in pottery, ironworking, agriculture, and the spread of glass beads and cowries, which traveled a thousand miles from places like Shanga or Kilwa.67 In that period, a very old pottery tradition called “Urewe,” ware turns up on the Inland Sea littoral. But the oldest Urewe pots—from the second quarter of the last millennium bce—have been found far to the west, in the highlands of Rwanda. They first turn up in littoral settings after 500 bce. They are well known among specialists for features that take potters lots of time and skill to make. Those features grew rougher after the 6th century ce, ahead of a dramatic shift in vessel sizes, assemblage, and decorative grammar that swept over the littoral. As these new ceramics, called Entebbe Wares by the archaeologist Ceri Ashley, grew ubiquitous by 1000 ce, the transitional Urewe wares grew scarce, eventually disappearing from littoral life. The new wares are only found in sites on or within easy walking distance (<7 kms) to the shore. This distinctive distribution, just like the distribution of the term musambwa, for territorial spirit, marks the emergence of an intercommunicating littoral world.

The new vessel sizes in the Entebbe ware assemblage—probably used for serving beer—were so large that people couldn’t move them when full of liquid. They represented a new scale in long-standing social practices of feasting and hospitality. Homesteads with the big beer pots attracted visitors. And, in using the new style for larger scales of social networking, people turned away from the aura of antiquity invoked by using Urewe ware pots, which had, after all, been around fifteen centuries, an extraordinarily long life for a single potting tradition, by the time Entebbe wares took over. This way of looking at material culture, like pottery, doesn’t treat it as a proxy for ethnic or linguistic identity; it looks for the inner worlds of ancient people, reflected in the ordinary things they made and used and in the organization of the domestic space they inhabited. Some of their things were extraordinary.

In 1929, a pressgang of African prisoners, clearing land for the expansion of the colonial prison in which they were incarcerated, uncovered a buried set of terracotta figures, recently dated to between the 10th and 12th centuries.68 One figure is a head with a coiled neck-ring, an icon of a python’s constricting method of crushing its prey (another metaphor, like being kicked, for the human experience of being possessed by a spirit). The other figures in the assemblage are headless, but they wear bangles on their wrists. The Luzira Group, as they are known, was found a few kilometers from the water. The Group invites analysis and interpretation like Ashley’s socialized approach to pottery. Giant beer pots implied social interplay, and these figures possess representations and depositional contexts pointing to the particular social interplay of possession by spirits. Documentary accounts from the 19th century tell us that pythons may house a musambwa spirit. Oral traditions tell of a littoral culture hero, Kintu, who helped establish what became the most powerful monarchy on the shores of the Inland Sea after he and his allies cut off the head of a powerful snake-ruler (see “Traditions, Written and Oral” and “New Directions: Conceptual Metaphor and Materiality; Words, Things, and Meanings Return”, below, for further discussion). The likes of the Luzira Group disappeared from the region’s traditions of plastic arts, perhaps as early as the turn of the first millennium, just ahead of the intimacies of 12th century climatic uncertainty. Its disappearance heralded a new era of littoral ethnic formation reflected by people innovating new terms for expansive clanship (ekika) and for portable territorial spirits (musambwa), both operations that were lubricated by the beer in the large, refillable pots.

Traditions, Written and Oral

Such intimate dimensions of the social history of ethnic formation or feasting or trade pointed to by archaeology, historical linguistics, and paleo-environmental studies, work explicitly in oral traditions about groups, such as clans, shrine managers, and royal dynasties. Oral traditions are “verbal messages which are reported statements from the past beyond the present generation.”69 Oral histories are statements about the present generation. Oral traditions tell migration itineraries, military conflicts, famines, and claims about the origins and fates of groups. They include names of places, people, animals, plants, and objects often related to claims about which groups were first in a place or a region. To get at the historical riches in oral traditions, one begins by understanding what shapes the transmission of its messages.70 Oral traditions were performances before they became written accounts. They contain evidence about both the construction of the past and the past itself. But, their messages often have many meanings. So, the phrase “oral traditions” can be misleading because they may be strongly influenced by material that performers have learned from books or other media. Tracking the influences of performance and writing on oral traditions is an example of understanding what shapes the transmission of the messages they convey. But, the influences of writing on the contents of performed oral traditions do not extinguish their value as evidence of past events and past political realities.

By analyzing oral traditions in comparison with other versions in the archival and published record, a rich body of sources emerges, bridging the historical linguistic and archaeological evidence to the first-hand written accounts that turn up in volume for the coasts, from the 1500s, far later for the interior.71 The chronological reach of oral traditions can be impressive, upon reflection. The earliest such accounts related to the Inland Sea appear in the writings of Europeans who arrived in the 1860s. (Though literate Muslims had reached the Inland Sea a generation earlier, they left no written accounts of their experiences there.) Some of the Europeans’ reports included first-hand accounts by Africans of events that occurred early in the 1800s. Those accounts contained memories of events experienced by others a generation or more earlier than the teller’s—the 1740s at the earliest. They were then narrated to youths, early in the 1800s, the same people Europeans encountered in the 1860s as elderly African men and some women.72

The potential chronological reach of oral traditions is important, but other analytical challenges must be met in order to interpret their historical content. Above all, one needs different versions to study comparatively. Clan traditions are widely found in Africa, and they may be treated in this way. A clan history is very often a history of a network told through a metaphor of descent that manages contests over the contributions of different nodes in the network by placing them at particular points in an unfolding story. The older or earlier figures in the traditions with which later groups enjoy a tie of affinity or marriage ennoble each other. Clan traditions—indeed, all oral traditions—unfold in a dynamic tension between teller(s) and audience. It is therefore dangerous to take literally the rhetoric of descent that shapes them. Audiences understood the generations traced across the land over time by a series of migrations as networks of political affiliation and social opportunity, creating and joining smaller, dispersed communities.73 Where royal and clan traditions coexist—as they often do, for royalty was a common feature in early Africa’s past—the latter tend to be calibrated by the shifting power of the royal center, as revealed in the different versions of events one finds in the two genres.74 If one studies performances at a royal capital or a chief’s compound and then studies them at a shrine or a place claimed by a particular clan network as an early home, the idiom of a clan’s oral tradition changes, departing from royal traditions in rich ways.75 Indeed, the messages one retrieves from oral traditions, when properly analyzed and interpreted, are nothing less than traces of earlier political realities. As with any other historical source, the scholar must weigh the effect on its contents of the political contexts of its production in order to know how to use it.

Two ruptures in the history of the northwestern region of the Inland Sea—especially the parts touched by the expansionist monarchy known as Buganda—shape reading the documentary record—including written forms of oral traditions—in important ways (see Map 3). Each stimulated intellectual work in an idiom of genealogy, a common consequence of rapid change and cultural interplay.76 The crystallization of evangelistic energies and the embrace of Islamic and Christian monotheism was the first break. It began in the 1840s, with royal engagement with mobile traders who were Muslim, and reached a turning point in 1888, when the Ganda king, Mwanga, executed a number of commoner and noble converts to Christianity. The curiosity royals showed these religious affiliations and the practices they entailed culminated in deaths that transformed tolerance into the destruction of older public healing practices. Many were driven underground, others were glossed anew with explicitly Christian meanings, and still others were defeated and destroyed.77

The second rupture was the creation of a market in private property, following the signing of the Buganda Agreement, in 1900. In the struggles over ownership of mailo (as the gazetted parcels, most of which were square miles, were known in Luganda) that ensued, disputants often drew on clan histories to advance claims of earlier, sometimes ancient presence in a particular territory.78 The argument bound estate, region, and clanship tightly together. Yet, the politics in this legalistic setting do not mean that specific arguments about such antiquities were fabricated out of thin air. There were limits to such invention. For the claims of priority or antiquity in a particular territory affected by mailo designations to stick, they had to be crafted with categories, figures, and place names already imbued with the verisimilitude of antiquity in the minds of the literate, non-literate, missionary, and court audiences such claims sought to reach.

Those histories produced kinds of clanship obsessed with struggles over claims to land, repeatedly selecting particular instances of their earlier history as grounds for the claim. Their choices limit what guild historians may analyze and interpret in such histories to their rhetorical structure and the figures or places named in them. These are just the things largely missing from archaeological and historical linguistic evidence.

Whatever past one hunts down with oral traditions, it is indispensable to grasp the ways in which culturally specific but shifting concepts of time, space, and truth operate in them. Space is defined geographically, as a region composed of nodes of authority, and relationally, as a set of links between those nodes. Descent logic marks the unfolding of time by assigning generations a basic referential kinship standing. Connecting leading figures in a generation with particular hills or trees binds time and space in culturally precise ways called chronotopes. Huge amounts of information are often compressed into the earliest and the most recent generations. The longer ago—or the earlier in the line of descent linking different generations—an event occurred or a persona existed, the greater the weight of truth behind it. But alternative versions use the same logic to dispute one another, by shifting the order of generations or by inserting new ones in the sequence or by leaving some out of the mix, or all three. Clan histories, then, tell about the shifting political affiliations of many differently sized groups, so it is unsurprising and valuable to find in them displays of dispute, conflict, and fission.

The modular qualities of genealogical reckoning facilitate its continual revision, an observation frequently used to disqualify such accounts from being considered records of the past. But not all units in such reckoning are of equal relevance in the politics of revision. The names of founder figures of lineages and clans, or the names of particular territorial spirits may be moved up and down the line of time—and, thus, up and down in the rank antiquity accords. Comparing variant traditions that take up the same topics to find in their agreement a message transmitted across the generations is not the only way to use such narratives as historical sources. Clearly, a single generation in the traditions is not to be taken literally as the productive life span of a single individual. It brackets a sequence (or cluster) of events in which its central figures were prominent actors. By placing the same sequences—including momentous changes—inside the same generations, in performances undertaken over the years, tellers and audience colluded to lend the coherence and familiarity of individual lives to the events and personalities involved in the sequence.

Oral traditions plausibly connected to a region of archaeological survey or to a set of sites in a region have rich benefits. They often contain names, titles, or sayings with metaphorical content that sometimes appears on material culture. For examples, versions of the Lungfish clan’s history agree that one Mubiru Gabunga traveled along the Inland Sea’s northern littoral and some of its island archipelagos, from east to west. Along the way, he established a network of smiths, canoe-builders, fishing groups, and spirit mediums, by stopping at locations where each kind of skilled person resided. When his generation’s travels were completed, he settled at a safe harbor, with the assistance of a couple of other clan networks in the region. There he relinquished leadership of the clan—by placing a copper bracelet on the wrist of a “son” called Ssematimba (father of the pythons), who then moved to a different harbor and stayed there.79 Recall that the Luzira Group included headless figures wearing bangles on their wrists. The generation of Mubiru Gabunga and his “brothers” had defined the clan network spatially and relationally. The next period or generation was opened by the copper-bangle wearing Ssematimba, who established authority over territorial spirits represented by the figure of the python. When set alongside the iconography of the terracotta Luzira group, the stakes of traditions and the messages in the assemblage enter a dynamic conversation about a new scale of politics.80

Recounting the itinerary of travels undertaken by founding generations of a clan, traditions craft a map that is both geographical and relational. The map defines a region of activity for the clan by linking nodes in it. Nodes are stopping places. Links are movements between them. The shapes of these regions may help make sense of facies in a pottery tradition, or the temporal overlap of two different pottery traditions, one fading away and the other coming to life, in the same region. The traditions often specify why the nodes and links mattered to people, pointing to what brought pottery traditions to life or led to their disappearance. Lungfish clan histories do not agree on every stop in this itinerary, but the stops on which they agree all lay very near the Inland Sea, or on one of the islands that dot its northern reaches (Map 3). The itineraries also agree that Lungfish clan founders and their followers moved by canoe. Of the several “brothers” said to have founded the clan, one of them is always recalled as a smith. Fishing in the Inland Sea was an important economic activity that differed from fishing in rivers or smaller lakes. It required specialized knowledge of fish ecology in a huge and diverse environment, technical skill in making and setting large nets, and skill in canoe-navigation and understanding weather patterns. Lungfish clan traditions trumpet that skill by dropping the name of a famous harbor or the name of a village wealthy with fishers and smiths.

Concise vernacular periodization for clan origins exists, expressing a hegemonic view of what mattered in the past.81 In the North Nyanza-speaking areas of Uganda, it defines four periods for clan origins by referencing two major culture heroes, Kintu and Kimera. Kintu found the first clans already in the region when he arrived. The next clans traveled with Kintu. After that, another set of clans came with Kimera. The balance of clans not accounted for in the first three periods constitutes the fourth, those created after Kimera’s time. With origins settled, clan histories next developed a geographical frame and substantive contents of their founder figures’ most important accomplishments, such as Ssematimba’s earning the brass bangle of leadership at a particular place on the littoral. Migratory itineraries draw the map. Names, situated genealogically, and place names, situated in the itineraries, point to the substance of clan history. A list of duties, entitlements, or conflicts relating to Ganda royals (mostly kings, sometimes queen-mothers and queen-sisters) follows. One does not ask about the calendar dates for the origins of clans out of respect for the intellectual contents of clan histories, the events, and processes of dispute and conflict they recount. The concision of the vernacular approach restrains historians from seeking a fine-grained chronology that exceeds the intellectual contents of the tradition makers who were arguably unconcerned with such things until the 19th century.

New Directions: Conceptual Metaphor and Materiality; Words, Things, and Meanings Return

As the foregoing has made abundantly clear, early African history is a rich domain that requires interdisciplinary work. A central question shared by those disciplines revolves around the ways in which meaning was made in the past and the ways in which scholars might or might not generate historical evidence of that process. Well-dated documents and objects, with clear provenience, allow the historian to recover the contexts in which they were used, grounding arguments about their meaning in a rich medium of social, intellectual, economic, and political life. The fact that each source reveals context differently, lending different kinds of intimacy to the histories they support, is a boon to historians, allowing them latitude to explore different scales of peoples’ lives. But, as we have seen, scholars of Africa’s early history struggle to keep material culture, language, ecology, and making meaning together, in the same time and place. The resulting palimpsest of contexts—some, like the settings for the Bantu Expansions, are strewn across many centuries and regions and others, like the feast at Vumba Kuu, occurred in a matter of hours or days in one part of one town on east Africa’s Indian Ocean coast—challenge historians to keep the meanings of the use of language or serving bowls alive at both scales.

One way to do so is to focus as much as possible on evidence for making meaning that reflects something about contexts of use. Where archaeologists found the serving bowls at Vumba Kuu is as important to understanding the meaning people used the serving bowls to make as is the kind of food the bowls might have contained, their number, and so forth. The idea of the “bush,” where metals were worked and animals or fish hunted was as important to producing fame and renown—qualities of meaning attached to action and knowledge—as were successful smelts and kills, in and around the central Zambezi basin, in the first millennium ce. The more we can say about the ways in which people worked with meaning in the past, the richer and more accessible will be Africa’s earlier histories.

Conceptual metaphors and their different iterations in material, spatial, and linguistic form, are a good way to do this. They provide insight into context, helping us understand the ways in which innovators got their innovations noticed, and pointing to other hard-to-access matters of communication and interaction between individuals available to most historians who work with well-dated documentary records, but notoriously difficult to reconstruct from the conventional kit of sources for Africa’s earlier histories. Metaphor is an agent for making meaning, its locus classicus in historical linguistics and a common theme in literary studies.82 But, the importance of metaphor that is of most interest, here, is more than a literary device for comparison. It is a dimension of human experience that people use to make meaning with objects and spatial form, not only with language.83

Metaphor makes meaning through juxtaposing some aspect of one thing (the source) in order to alter the understanding of another thing (the target). The altered understanding is the new meaning we’re after, and the choice of the source might reveal something about the context through which people proposed the new meaning. We are most familiar with spoken or poetic metaphors: “I fell through a trapdoor of sorrow” or with written forms in famous texts like those of Shakespeare: “the slings and arrows of outrageous fortune.”84 In both examples, qualities of the source—a trapdoor, the tools of ballistics—shed new light on the target, which is often something more abstract—like sorrow or fortune. The juxtaposition proposes new or particular attention to the abstract domain of meaning. Sorrow becomes an edgeless expanse, just beneath us. Fortune brings with it more than pleasures and comforts.

The juxtapositions work—they convey the new or particular meaning—by drawing on features of our experience. We know from past experience the dangers of falling. We know from past experience that bad can come with good, that intended outcomes might have unintended consequences. So, a metaphor concentrates the work of understanding. It can expand literal, taken-for-granted meanings into new domains, or it can use well-established features of experience to pin down elusive qualities of experience and thought. Well-chosen metaphors enhance communicative process by making meaning that is accessible. And, the choices reflect the quarry of historically specific meaning making: something of the contexts in which it (the choice) was made.

Before exploring the ways in which that last point works, we need to dwell a little further on processes of meaning making to appreciate its multiplicities and see the ways in which a well-chosen metaphor can focus different understandings without reducing their diversity. So many variables of social position, gender, age, power—in a word, context—shape making meaning that scholars might approach it like guild historians do: as historically specific literal meaning. Literal meanings are “basic assumptions about what signs are and how they function in the world” and “what kinds of agentive subjects and acted-upon objects might be found in the world.”85 In other words, literal meanings are only obvious—that is, literal—to people sharing these assumptions.86 Keane’s approach frames meaning as action. He asks us to think about what something can mean and not to think only about stable fields of meaning mapped by reconstructed words located in particular steps (proto-languages and their vocabularies) in a sequence of language divergence. Keane draws our attention to the persistent need for reference in the world—that is, the persistent need to make oneself understood by others—and the excess of interpretation—excess, because not all interpretation is germane to a particular instance of communicative understanding—that the need to communicate effectively generates. Conceptual metaphors possess ordering properties that assist in creating knowledge and conveying it. Conceptual metaphors channel understanding in particular ways, reducing confusion. In a word, conceptual metaphor enhances clarity in communication, in making meaning.

But conceptual metaphors are layered; they are not first of all linguistic forms. That is, ironically, a main source of confusion about them. I say “ironically” because, in speech, people use metaphors to reduce confusion and convey clarity. Primary conceptual metaphors, as cognitive linguists call them, are image-schemas or maps, drawn from embodied experience. They structure the conceptualizations that motivate linguistic or material metaphorical expressions.87 For example, the conceptualization of an angry person as a container under pressure has a basis in the experience of one’s body temperature rising along with intensifying emotions. The simultaneous unfolding of those associations in embodied experience is structured. This configures elaborations, such as the sense of containment of the heat and emotion generated by building pressure, raising questions about how the pressure builds—is the container heated or not?—and “what kind of substance fills the container (liquid, substance, or objects), what consequences the explosion has.”88 Such metaphor mapping is often filled out in culturally and historically specific ways. The specifics take linguistic form in compound words, in a word’s polysemy, and in seriated semantic shifts all of which may be placed in a sequence of speech community events such as divergence and transfer.89 They may also have iterations as phrases in oral traditions. Some have iterations in decoration, form, spatial organization, and in the use of and depositional practice with objects, like pots. We work with these so-called cultural conceptual metaphors in our daily lives in exploring or using what a word, a thing, or a spatial arrangement can mean.

But, using bodily experience as a source for thinking about abstract targets in new or clearer ways is not the only way in which metaphor and concept interact productively. Conceptual blends occur when something in the target did not exist in the source until someone improvised that relation in an expression, an action, a thing, or all three, giving rise to novelty.90 A classic example appears in the sentence “The fence runs all the way down to the river.” As Edwin Hutchins explains, the sentence blends stasis and motion. A fence and a river’s edge are blended with “a shape of motion (movement down a hill to the edge of a river) along the relevant dimension of a static object (the length of the fence).” This is a blend because the emergent entity (the way to look at the fence, as “running”) is not present in either the source or the target but is a combination of an element of the source (the fence’s long shape) with an element of the target (the fact that one runs from point A to point B, in a line, but does so in a temporal sequence). The “fictive motion” accorded to the fence “blends the shape of a movement through space with the spatial shape of an object to produce a temporal sequence of attention to the shape of the object.”91 The point here lies in the complexities of meaning making that conceptual metaphors make possible.

The Inland Sea historical thread spun here includes an innovative conceptual blend. The improvisation lay in proposing that mediums were pythons. Mediums and priests did this by combining a supplicant’s experience of a possessed medium (who was not a python, but whom supplicants understood, by the medium’s behavior, to be possessed by a territorial spirit [musambwa]) with the experience of pythons that supplicants understood were territorial musambwa spirits (but not mediums). Its lexical form, èmmandwà, is distributed widely in Rutara and North Nyanza languages, where it may refer both to a medium and a spirit. Only the Ganda language has the additional gloss, “python.” The proposition also took material form in the depiction of a ring coiling like a python around the neck of the Luzira Head. The proposition created the concept of a spirit medium that is a python. Its lexical and material iterations allow us to date that proposition to around a millennium ago.92

This example reveals the importance of marking in improvising and innovating meaning with words and things. Marking reveals the moment when particular concepts emerged. When North Nyanza-speakers marked èkâ (homestead; at home) with the èkì- prefix, producing èkìkâ, (clan), they proposed a new conceptual frame for a kind of group, suggesting it be understood, in part, as an analog to peoples’ experiences of homestead dynamics.93 Speakers of Luganda, one of the languages that formed after Proto-North Nyanza ceased to exist as a single intercommunicating speech community, drew on that conceptual frame to elaborate, through naming, smaller-sized groups comprising an èkìkâ. An èssigà or an òmùtuba are often translated into English as [lineage]. But each word is polysemous. An èssigà is one of the three stones forming the hearth on which women prepare meals in a homestead. An òmùtuba is a kind of ficus tree, whose bark is used to make cloth. Such trees are ubiquitous in homesteads. Another word often given the English translation of [lineage] is òlùnyiriri. It departs from the entailments of categorization given by the homestead framing because it is also a word for [line] or [row] or [verse]. The departure reveals that people were not blindly hemmed in by the conceptual framing of clanship. They were free to insert other ideas such as, in this case, the importance of reckoning descent through a line or row of generations. But, the homestead frame is also implicated in naming the very smallest parts of clans, the parts formed by individual extended families, usually three generations deep. An òluggyà and an ènjû name these smallest units of a clan. The òluggyà is also the courtyard outside the house, and the ènjû is the structure of the house itself.

Situating lexical and material iterations of these ideas and things in the flow of time reveals the moments or contexts in which conceptual entrepreneurs worked. Before taking on a hegemonic, habitual role in communication—as people did with èmmandwà or èkìkâ (and its elaborations, just discussed), which both remain in Ganda usage—the juxtapositions proposed in a new metaphor, or an elaboration of an old one, stood out from other, literal meanings. The new meanings on offer through juxtaposition were ideological; they were marked. Yet, the durability of the idea or object in question—its having continued to be spoken or made—was accomplished by making it literal through repeated use in the untold social interactions that kept it alive through active elaboration. A stylized python, with a clay-head and a raffia-cloth body was used in the 17th century to extract confessions from royal prisoners held by the Ganda state.94 This is an example of such durability, an effect of elaboration. Butsome, like the Luzira head, fell out of use, or grew stale, no longer generating continuing elaborations and blends. Such variations reflect the outcomes of the contests that lie behind the production of habit and literal meaning. They are basic, intimate, intellectual, and material changes over time, the grist of any historian’s mill.

As described, juxtaposing elevated body temperature to emotional states, or juxtaposing experiential knowledge of python behavior to that of a medium possessed by a spirit, draws on experience as a source of understanding and conceptualization. Perceptual processes are widely used as sources for abstract concepts of knowing and feeling. “Seeing is believing” expresses that relation in English, but Eve Sweetser has shown that juxtaposition to have a great time-depth, reaching back thousands of years to the time when the Indo-European speech community existed.95 “I hear the fever,” a literal translation into English of an expression very widespread in Africa’s Bantu languages, proposes that bodily noise is the way to knowing what’s going on inside or with one’s body. Clearly, perceptual processes are important sources for conceptualizing knowledge. Their everydayness lends a hegemonic or habitual or literal quality to the juxtapositions of which they are a part. Repetitive, widely shared bodily processes (including the perceptual ones just mentioned) sustain this quality of literalness.

Metaphor and blending are not only tools for expanding the ability of historians and archaeologists to think through, without contemporary documents, the ways in which people made meaning in the past. They also touch on particular themes of great interest to scholars and students of Africa’s recent past as well. Let us not forget that, when first proposed, a new relation between source and target stands out; it marks a new understanding on offer. These twin qualities of metaphor—marking novelty and becoming literal or taken for granted—open up important dimensions to the argument that we take seriously the content of primordialist claims meant to instigate ethnic identification. Appeals to shared ancestry, distinctive cuisine, or styles of speaking as concepts or media of belonging, touch on intimacies of thought and action. The historian Jonathon Glassman suggests their appeal has something to do with their intimate links to everyday practices and rituals.96 The same banalities of embodied experience of the world that recommend it as a source for making conceptual sense of abstract targets like belief, through metaphorical juxtaposition of the source to the target, recommend them to the study of primordialist appeals in racial, ethnic, and nationalist thought.97

The structuring properties working in such juxtapositions worry those who find meaning a much more elusive, contingent quality of social life than the orderly principles of conceptual metaphor imply.98 For them, cultural understandings worked out in contexts of language use, not the entailments of a particular metaphor, produce meaning. But, are the two mutually exclusive? The structuring properties of conceptual metaphor are habits of abstract thinking, they set broad limits for the cultural work of elaborating them in particular directions of meaning. The details of the elaborations of meaning reflect the contingencies of context but their shapes are partly set by the possibilities implicit in the structure of the superintendent juxtaposition making the metaphor. The details of meaning worked out in communication often elude historians of early Africa, who work without written documents. Conceptual metaphor and blending can bridge universally embodied perceptual experience to the enormous variety of elaborations and understanding prompted by and constitutive of power, scale, and culture.99 Because they may have lexical and material iterations, they cross historical linguistics, oral traditions, and archaeology. Because they work the grain of embodied experience to invent new ways of being in the world, they cross hegemony and ideology. They represent African history on its own terms as well as facilitating its comparison with any other history.

Africa’s Early History: Neither Traditional Nor Exotic, But Inspirational

Historians have developed a conventional approach to writing Africa’s earlier history in which they take interpretations of each source discussed here and play them off against each other, to confirm, deny, or revise their conclusions. The result conforms to disciplinary conventions for the critical analysis of empirical evidence, revealing Africans coming up with familiar and novel solutions to geographical, economic, political, and intellectual challenges. Africa’s early history has never been traditional because Africans, like anyone else, improvised and innovated in the face of new circumstances and aspirations.100 It is a place with a past whose time has come: to provincialize a forgetful globalism in which Africa is a place of relentless localness when, in reality, it was never only a collection of localities; they were networked within and beyond the continent’s physical edges, and the networks reflected the interests of continental people, not only of others from elsewhere. By renewing the importance of the terrestrial, of rivers, lakes, seas, and oceans, as zones of mobility, Africa’s early history reveals enormous creativity and success in the art of living. Telling those stories relies on the conventional set of sources explored here, each with different strengths and weaknesses. Time and region shrink and expand differently, actors, ideas, feelings, food, home, and movement come in and out of view differently, for each source. Like more familiar historical sources, they leave us with an interpretive palimpsest of traces of the richness of life in the past. When they converge, each from its different angle, on social facts like clanship, feasting, bushcraft, or ethnic formation, their resolution pushes the muddle of the palimpsest aside, like when the shards of glass in a kaleidoscope snap into patterns, one after another.

Discussion of the Literature

Scholars have tackled the challenges of writing the history of times and place beyond literacy by using all of the different kinds of sources discussed. They have been at it for some time. In 1916, the American linguist and ethnographer, Edward Sapir, published an early manifesto on these methodological challenges.101 Keen to promote interdisciplinary work, this foundational publication had a lot to say about blending evidence from language, material culture, and geography, but said little about documents, oral history and oral tradition, archaeology, and physical anthropology. Africanists, however, took up these challenges, beginning in the colonial period and continuing, today.102 Many others, working all over the world—not just in Africa—have answered Sapir’s call and developed historical and anthropological methods together.103 The daunting challenges of mastering the specialized fields to be integrated in such an approach have, perhaps, limited its growth.

This approach has been used to examine questions about the past that have changed profoundly over the generations since Sapir wrote. Once basic matters of chronology and historical geography are resolved, scholars want to know more about people’s social lives, what they valued, the technologies they developed, their relations to their environment, to their past, and so forth. Humanistic concerns, then, have pushed methodological innovation, including, for example, the application of anthropological genetics to distinguish migration from diffusion or the application of theories of meaning making to bring communicative settings to life even without direct evidence of communicative acts. The resulting histories put people’s actions and aspirations at the center of accounts. As scholars continue to explore the constraints—such as environmental conditions and agro-economic profiles—that shaped human action in the past, they confront the limits such a focus imposes on their work. People are surprising—at times content with routine, at other times restlessly improvising solutions to immediate problems, with intended or unintended consequences for their routines. This mix of innovation and conservatism produces continuity and change recognizable at the long time scales and vast reaches of geography that so many scholars of early Africa’s past work with.

The strengths and weaknesses of such interdisciplinary work shape the many research questions that remain open today. Three important issues have been explored in this article: the social and intellectual history of the Bantu Expansions; the shape of intellectual debates and organizational structures (and the factors conditioning them) of earlier African political cultures; and the social basis of supply and demand in African economies. The first one requires integrating the conditioning forces of ecological process, food systems, and technology with cultural and intellectual work. The second one puts ideological conflict and bureaucratic culture into the study of institutions, exchange, and warfare in thinking historically about African politics. The last one pushes decisively away from ahistorical assumptions about the determining force of the aggregate of atomized, self-interested actors in driving economic life. In place of Homo economicus, we now study the cultural and historically specific factors—like social responsibility or ideas of value—shaping people’s decisions to produce, exchange, and use things. Historians of early Africa share all of these moves with historians of other times and places, whether beyond literacy or not.104

The new questions being asked about old topics like mobility, state formation, and economic change, reflect broader trends in the academy as well as the politics of knowledge in Africa. At a regional level, they include some very localized dynamics, such as the significance of archaeology in Rwanda for thinking about recent violence, or the need in South Africa and elsewhere for new approaches to the vast period called precolonial.105 What are the contours of early African histories of violence, healing and harming, gender, inequality, or ethnicity? What do they say—if anything—about other histories of those topics, and do such histories inform in new ways modernist approaches to the topics in Africa today? In what ways can and must we broaden imperial and colonial concepts and experiences of race and ethnicity when writing Africa’s earlier history? When must we replace them with those whose roots in the earlier past allow us to critique more recent (and, perhaps, familiar) forms?106

The pace of basic research into Africa’s earlier history is slow for a variety of reasons. Working conditions in some regions are difficult. The benefits of digitization are uneven, although historical linguists are benefiting a lot, so far.107 The prospects for collaboration in teams of disciplinary and area studies specialists in writing early African history depend on funding. This makes Africa-based teams especially rare, given that the risks of national politics influence what counts as basic research in the past of a given country.108 South Africa, Nigeria, and Kenya, to name a few countries with considerable university infrastructure, are best positioned to pursue this tack. Researchers based in other African countries have developed their own transnational networks, as access to support for research and writing leave is now reaching the junior ranks of Africa’s faculty.109 In many settings, collaboration is ad hoc, depending on the culture of networking in the constituent disciplines. Most often it amounts to sharing of basic research across disciplines rather than working on a single research problem defined collaboratively. Archaeologists are far more used to working in interdisciplinary teams, the diversity of which is normally a function of funding. Historians and cultural anthropologists, on the other hand, often stay close to their archives and the places and people or field sites they know best; but they read the secondary literature in disciplines adjacent to the problem they are examining.

Primary Sources

Given the technical skills required to analyze and interpret the evidence from the sources discussed in this article—and compare the findings critically—a reader without those skills might find the evidence, in its raw forms, hard to assess and use. In that case, curious readers should consult the case studies cited in the footnotes and bibliography. Otherwise, one might consult the following sources. Ancient environmental and ecological change tends to be divided by community (plant, pollen, fish, and so forth) or to focus on agriculture. For pollen studies, see the Observatoire Midi-Pyrénées. For historical linguistics, see Christopher Ehret, A Historical-Comparative Reconstruction of Nilo-Saharan; Christopher Ehret, Reconstructing Proto-Afroasiatic (Proto-Afrasian): Vowels, Tone, Consonants, and Vocabulary. For Bantu, see Yvonne Bastin, André Coupez, Evariste Mumba, and Thilo C. Schadeberg, Bantu Lexical Reconstructions 3. As a partial guide to the latter, see Koen Bostoen and Yvonne Bastin, “Bantu Lexical Reconstruction.” On lexicostatistics, see George Starostin and Phil Krylov, The Global Lexicostatistics Database. The Global Lexicostatistics Database website contains information on lexical reconstruction in Berber (North and Saharan Africa) and KhoeSan (Southern and Eastern Africa). For archaeological site reports, see among others, British International Archaeological Reports (publisher and distributor). Though not focused on Africa, one can also find primary datasets at the Archaeology Data Service website, including material on east African coastal pottery. For a rough start on oral traditions, consult the International Society for the Study of Oral Traditions’ website.110

Further Reading

  • Blier, Suzanne Preston. Art and Risk in Ancient Yoruba: Ife history, power, and identity c. 1300. Cambridge, UK: Cambridge University Press, 2015.
  • Bostoen, Koen, Bernard Clist, Charles Doumenge, Rebecca Grollemund, Jean-Marie Hombert, Joseph Koni Muluwa, et al. “Middle to Late Holocene Paleoclimatic Change and the Early Bantu Expansion in the Rain Forests of Western Central Africa.” Current Anthropology 56, no. 3 (2015): 354–384.
  • Chami, Felix. “The Problem of Equifinality in Archaeology.” In Africa in Theory, Theory in Africa: Locating Meaning in Archaeology. Edited by Stephanie Wynne-Jones and Jeffrey B. Fleisher, 38–47. London: Routledge, 2015.
  • Chirikure, Shadreck, M. Manyanga, Innocent Pikirayi, and M. Pollard, “New Pathways of Sociopolitical Complexity in Southern Africa,” African Archaeological Review 30, no. 4 (2012): 339–366.
  • de Luna, Kathryn. Collecting Food, Cultivating People: Subsistence and Society in Central Africa. New Haven, CT: Yale University Press, 2016.
  • de Schryver, Gilles-Maurice, Rebecca Grollemund, Simon Branford, and Koen Bostoen. “Introducing a State-of-the-Art Phylogenetic Classification of the Kikongo Language Cluster.” Africana Linguistica 21 (2015): 87–162.
  • Ehret, Christopher. An African Classical Age: Eastern and Southern Africa in World History, 1000 BC to AD 400. Charlottesville, VA: University of Virginia Press, 1998.
  • Ehret, Christopher. History and the Testimony of Language. Berkeley: University of California Press, 2011.
  • Farias, Paulo Fernando de Moraes. Arabic Medieval Inscriptions from the Republic of Mali: Epigraphy, Chronicles, and Songhay-Tuareg History, Fontes Historiae Africanae, new series. Oxford: Oxford University Press, 2003.
  • Fields-Black, Edda. Deep Roots: Rice Farmers in West Africa and the African Diaspora. Bloomington: Indiana University Press, 2008.
  • Grollemund, Rebecca, Simon Branford, Koen Bostoen, Andrew Meade, Chris Venditti, and Mark Pagel. “Bantu Expansion Shows that Habitat Alters the Route and Pace of Human Dispersals,” Proceedings of the National Academy of Sciences 112, no. 43 (2015): 13296–13301.
  • Hamilton, Carolyn, Bernard K. Mbenga, and Robert Ross (Eds.) Cambridge History of South Africa, Vol. 1: From Early Times to 1885, Cambridge, UK: Cambridge University Press, 2010.
  • Haour, Anne (2013). Outsiders and Strangers: An Archaeology of Liminality in West Africa. (Oxford: Oxford University Press, 2013).
  • Klieman, Kairn. “‘The Pygmies Were Our Compass’: Bantu and Batwa in the History of West-Central Africa, Early Times to ca. 1900 CE. Portsmouth, NH: Heinemann Publishers, 2003.
  • Kodesh, Neil. Beyond the Royal Gaze: Clanship and Public Healing in Buganda. Charlottesville: University of Virginia Press, 2010.
  • Kropp, Dakubu, and Mary Esther. The Languages of Ghana. New York: Kegan Paul International, 1988.
  • Livingstone Smith, Alexandre, Els Cornelissen, Olivier P. Gosselain, and Scott MacEachern, eds. Field Manual for African Archaeology. Tervuren, Belgium: Royal Museum for Central Africa, 2017).
  • McIntosh, Roderick J. Peoples of the Middle Niger: Island of Gold. Malden, MA: Blackwell, 1998.
  • McIntosh, Susan Keech, ed. Beyond Chiefdoms. Cambridge, UK: Cambridge University Press, 1999.
  • Mitchell, Peter, and Paul J. Lane, eds. The Oxford Handbook of African Archaeology. Oxford: Oxford University Press, 2013.
  • Namono, Catherine. “Resolving the Authorship of the Geometric Rock Art of Uganda.” Journal of African Archaeology 8 (2010): 239–257.
  • Ngomanda, Alfred, A. Chepstow-Lusty, M. Makaya, C. Favier, P. Schevin, J. Maley, et al. “Western Equatorial African Forest-Savanna Mosaics: A Legacy of Late Holocene Climatic Change?” Climate of the Past 5, no. 4 (2009): 647–659.
  • Nurse, Derek. “The Contributions of Linguistics to the Study of History in Africa.” Journal of African History 38, no. 3 (1997): 359–91.
  • Ogundiran, Akinwumi, “Of Small Things Remembered: Beads, Cowries, and Cultural Translations of the Atlantic Experience in Yorubaland.” International Journal of African Historical Studies 35 (2002), 427–457.
  • Pikirayi, Innocent. The Zimbabwe Culture: Origins and Decline of Southern Zambezian States. Walnut Creek, CA: AltaMira, 2001.
  • Schoenbrun, David. “Mixing, Moving, Making, Meaning: Possible Futures for the Distant Past.” African Archaeological Review 29, no. 2 (2012): 297–300.
  • Schoenbrun, David. “Pythons Worked: Constellating Communities of Practice with Conceptual Metaphor in Northern Lake Victoria, ca. A.D. 800 to 1200.” In Knowledge in Motion: Constellations of Learning Across Time and Place. Edited by Andrew P. Roddick and Ann B. Stahl, 216–246. Tucson: University of Arizona Press, 2016.
  • Stahl, Ann Brower. Making History in Banda: Anthropological Visions of Africa’s Past. Cambridge, UK: Cambridge University Press, 2001.
  • Thiaw, Ibrahima. Espace, culture matérielle et identities en Senegambie. Dakar, Senegal: CODESRIA, 2010.
  • Vansina, Jan. Oral Tradition as History. Madison: University of Wisconsin Press, 1985.
  • Vansina, Jan. How Societies Are Born: Governance in West Central Africa before 1600. Charlottesville: University of Virginia Press, 2004.
  • Verschuren, Dirk, and Dan J. Charman. “Latitudinal Linkages in Late Holocene Moisture-Balance Variation.” In Natural Climate Variability and Global Warming: A Holocene Perspective. Edited by Richard W. Battarbee and Heather A. Binney, 189–231. New York: John Wiley Publishers, 2008.
  • Wynne-Jones, Stephanie, and Adria Laviolette, eds. The Swahili World. New York: Routledge, 2018.


  • 1. The article says little on anthropological genetics, but see Shomarka O. Y. Keita, “Physical Anthropology and African History,” in Writing African History, ed. John Edward Philips (Rochester: University of Rochester Press, 2005), 112–152; David L. Schoenbrun, “Mixing, Moving, Making, Meaning: Possible Futures for the Distant African Past,” African Archaeological Review 29, no. 2 (2012): 293–317, here 300–303; Scott MacEachern, “Genetics and Archaeology,” in Oxford Handbook of African Archaeology, eds. Peter Mitchell and Paul J. Lane (Oxford: Oxford University Press, 2013), 65–76, for constructive critique; Cesare de Filippo, Koen Bostoen, Mark Stoneking, and Brigitte Pakendorf, “Bringing Together Linguistic and Genetic Evidence to Test the Bantu Expansion,” Proceedings of the Royal Society B: Biological Sciences 279 (August 2012): 3256–3263; and Brigitte, Pakendorf, H. Gunnink, Bonnie Sands, and Koen Bostoen, “Prehistoric Bantu-Khoisan Language Contact: A Cross-Disciplinary Approach,” Language Development and Change 7, no. 1 (2017): 1–46, both cautiously positive.

  • 2. This article also says little about these documentary materials since they are taken up elsewhere in the volume. But see Anonymous, The Periplus Maris Erythraei: Text with Introduction, Translation, and Commentary by Lionel Casson (Princeton, NJ: Princeton University Press, 1989); G. S. P. Freeman-Grenville, ed., The East African Coast: Select Documents from the First to the Earlier Nineteenth Century (Oxford: Clarendon Press, 1962); Malyn D. D. Newitt, ed., The Portuguese in West Africa, 1415–1670: A Documentary History (Cambridge, UK: Cambridge University Press, 2010); George McCall Theal, ed., Records of South-Eastern Africa: Collected in Various Libraries and Archive Departments in Europe, 9 vols. (Cape Town: Government of the Cape Colony, 1898–1903); Richard Pankhurst, ed., The Ethiopian Royal Chronicles (Addis Ababa: Oxford University Press, 1967); John Thornton’s African Texts, translations of key texts by missionaries in West Central Africa from the 16th century; see also the general introduction by John K. Thornton, “European Documents and African History,” in Philips, Writing African History, 254–265. A good way into West African examples oriented to the Atlantic littoral and interior is Alice Bellagamba, Sandra E. Greene, and Martin Klein, eds., African Voices on Slavery and the Slave Trade, 2 vols. (Cambridge, UK: Cambridge University Press, 2016). Good introductions to Arabic manuscripts in translation include Nehemia Levtzion and J. F. P. Hopkins, eds., A Corpus of Early Arabic Sources for West African History (Cambridge, UK: Cambridge University Press, 1985); John O. Hunwick, Timbuktu and the Songhay Empire: Al-Sa‘dī’s Ta’rīkh Al-sūdān down to 1613 and other Contemporary Documents (Leiden, Netherlands: Brill, 1999); Multiple editors, Arabic Literature of Africa, 5 vols. (Leiden, Netherlands: Brill, 1993–2016); and Paulo Fernando de Moraes Farias, Arabic Medieval Inscriptions from the Republic of Mali: Epigraphy, Chronicles, and Songhay-Tuareg History, Fontes Historiae Africanae, new series (Oxford: Oxford University Press, 2003). This series publishes other important primary sources, both documentary and oral historical and/or oral tradition.

  • 3. James Fairhead and Melissa Leach, Misreading the African Landscape: Society and Ecology in a Forest-Savanna Mosaic (Cambridge, UK: Cambridge University Press, 1996); Daniel B. Botkin, Discordant Harmonies: A New Ecology for the Twenty-First Century (New York: Oxford University Press, 1990); and Bill Gammage, The Biggest Estate on Earth: How Aborigines Made Australia (London: Allen & Unwin Publishers, 2011).

  • 4. Timothy C. Weiskel, “Agents of Empire: Steps Toward an Ecology of Imperialism,” Environmental Review 11, no. 4 (1987): 275–288.

  • 5. For West African examples, see Rod McIntosh, Joseph Tainter, and Susan Keech McIntosh, eds., The Way the Wind Blows: Climate Change, History, and Human Action (New York: Columbia University Press, 2000); for Central Africa, see Richard Oslisly, Lee White, Ilham Bentaleb, Charly Favier, Michel Fontugne, Jean-Francois Gillet et al., “Climatic and Cultural Changes in the West Congo Basin Forests over the Past 5000 years,” Philosophical Transactions of the Royal Society of London B; Biological Sciences 368, no. 1625 (September 2013): 1–10; for northeast, east, and southern Africa, see James McCann, Green Land, Brown Land, Black Land: An Environmental History of Africa (Portsmouth, NH: Heinemann Publishers, 1999); and Peter Schmidt, “Archaeological Views on a History Landscape Change in East Africa,” Journal of African History 38, no. 4 (1997): 393–421.

  • 6. James M. Russell and Thomas C. Johnson, “A High-Resolution Geochemical Record from Lake Edward, Uganda-Congo and the Timing and Causes of Tropical African Drought during the Late Holocene” Quaternary Science Reviews 24 (2005): 1375–1389, here 1382; and Dirk Verschuren and Dan J. Charman, “Latitudinal Linkages in Late Holocene Moisture-Balance Variation,” in Natural Climate Variability and Global Warming: A Holocene Perspective, eds. Richard W. Battarbee and Heather A. Binney (Oxford: Blackwell Publishing, 2008), 189–231, here 211, fig. 8.11.

  • 7. Russell and Johnson, “High-Resolution,” 1382; Dirk Verschuren, “Decadal and Century-Scale Climate Variability in Tropical Africa during the Past 2000 years,” in Past Climate Variability Through Europe and Africa, eds. Richard W. Battarbee, Françoise Gasse, and Catherine E. Stickley (Amsterdam: Elsevier, 2004), 139–158, here 147; and Verschuren and Charman, “Latitudinal Linkages,” 211–212.

  • 8. David B. Ryves et al., Immaculate Ssemmanda, “Environmental Change over the Last Millennium Recorded in Two Contrasting Crater Lakes in Western Uganda, eastern Africa (Lakes Kasenda and Wandakara),” Quaternary Science Reviews 30 (2011): 555–569, here 567; but see David Taylor, Peter Robertshaw, and Rob A. Marchant, “Environmental Change and Political-Economic Upheaval in Precolonial Uganda,” The Holocene 10, 4 (2000), Table 1, p. 533; and Verschuren and Charman, “Latitudinal Linkages,” 189–231, esp. 210–214.

  • 9. J. Curt Stager, Brian Cumming, and L. David Meeker, “A 10,000-Year High-Resolution Diatom Record from Pilkington Bay, Lake Victoria, East Africa,” Quaternary Research 59 (2003): 172–181, here items “g” and “h” in fig. 7, pp. 179 and 180 (on seasonality).

  • 10. J. Curt Stager et al., “Solar Variability and the Levels of Lake Victoria, East Africa, during the Last Millennium,” Journal of Paleolimnology 33 (2005): 243–251, here fig. 4.d., 248.

  • 11. Koen Bostoen et al., “Middle to Late Holocene Paleoclimatic Change and the Early Bantu Expansion in the Rain Forests of Western Central Africa,” Current Anthropology 56, no. 3 (2015): 354–384, here 355–358; and Rebecca Grollemund et al., “Bantu Expansion Shows that Habitat Alters the Route and Pace of Human Dispersals,” Proceedings of the National Academy of Sciences 112, no. 43 (2015): 13296–13301; Alfred Ngomanda, Katharina Neumann, Astrid Schweizer, and Jean Maley. “Seasonality Change and the Third Millennium BP Rainforest Crisis in Southern Cameroon,” Quaternary Research 71 (2009): 307–318, here 313–316. A treeless grassland savanna spread along the coast and on the Bateke Plateau, far to the east.

  • 12. Alfred Ngomanda et al., “Western Equatorial African Forest-Savanna Mosaics: A Legacy of Late Holocene Climatic Change?” Climate of the Past 5, no. 4 (2009): 647–659, here 647–648, 655–656.

  • 13. Grollemund, et al., “Bantu Expansion”; Yvonne Bastin, André Coupez, Evariste Mumba, and Thilo C. Schadeberg, Bantu Lexical Reconstructions 3, Tervuren, Belgium, the Royal Museum for Central Africa now has some 10,000 entries. See also, Christopher Ehret, An African Classical Age: Eastern and Southern Africa in World History, 1000 BC to AD 400 (Charlottesville: University of Virginia Press, 1998); and Jan Vansina, Paths in the Rainforest: Towards a History of Political Tradition in Equatorial Africa (Madison: University of Wisconsin Press, 1990).

  • 14. Judith T. Irvine, “Subjected Words: African Linguistics and the Colonial Encounter,” Language and Communication 28 (2008): 323–343; Axel Fleisch, “The Reconstruction of Lexical Semantics in Bantu,” Sprache und Geschichte in Afrika 19 (2008): 67–106; and Sinfree Makoni, “Sociolinguistics, Colonial and Postcolonial: An Integrationist Perspective,” Language Sciences 33 (2011): 680–688.

  • 15. Derek Nurse, “The Contributions of Linguistics to the Study of History in Africa.” Journal of African History 38, no. 3 (1997): 359–391, here 361–362; Christopher Ehret, History and the Testimony of Language (Berkeley: University of California Press, 2011), 27–29. Examples include: Mary Esther Kropp Dakubu, Languages of Ghana (London: Kegan Paul International, 1988); Kathryn de Luna, “Classifying Botatwe: M60 Languages and the Settlement Chronology of South-Central Africa,” Africana Linguistica 16 (2010): 65–96; Malin Petzell and Harald Hammarström, “Grammatical and Lexical Comparison of the Greater Ruvu Bantu Languages,” Nordic Journal of African Studies 22, no. 3 (2013): 129–157; Derek Nurse, “The Diachronic Background to the Language Communities of Southwestern Tanzania,” Sprache und Geschichte in Afrika 9 (1988): 15–115; and Andrea Felber Seligman, “Determining Value: Southern Tanzanians’ Approaches to Trade, c. 300–1600 CE.” Unpublished PhD dissertation, Northwestern University, 2014; Gilles-Maurice de Schryver et al., “Introducing a State-of-the-Art Phylogenetic Classification of the Kikongo Language Cluster,” Africana Linguistica 21 (2015): 87–162; and Thomas Hinnebusch, Derek Nurse, and Martin Mould, Studies in the Classification of Eastern Bantu Languages (Hamburg: Helmut Buske, 1981).

  • 16. In corpus-based historical linguistics, the comparative method operates with documented vernacular forms, rarely reaching beyond the 17th century; cf. Koen Bostoen and Gilles-Maurice de Schryver, “Linguistic Innovation, Political Centralization, and Economic Integration in the Kongo Kingdom: Reconstructing the Spread of Prefix Reduction,” Diachronica 32, no. 2 (2015): 139–185, here 140–142.

  • 17. Called “similarity” or “distance” and “character” data, respectively, in recent scholarship; de Schryver, et al., “Kikongo Language Cluster,” 119–120.

  • 18. Henry R. T. Muzale, “A Reconstruction of the Proto-Rutara Tense-Aspect System.” Unpublished PhD dissertation, Memorial University of Newfoundland, 1998; and Petzell and Hammarström “Greater Ruvu.”

  • 19. Ehret, History and Testimony or Language, 105–132.

  • 20. Jan Vansina, How Societies Are Born: Governance in West Central Africa before 1600 (Charlottesville: University of Virginia Press, 2004), 13.

  • 21. Kathryn de Luna, “Surveying the Boundaries of Historical Linguistics and Archaeology: Early Settlement in South-Central Africa,” African Archaeological Review 29, no. 3 (2012): 209–251.

  • 22. Excellent examples, set in the Bandiagara region and in West Central Africa, appear on Ounjougou, Human Population and Palaeoenvironment in Africa (Bandiagara) and KongoKing (West Central Africa).

  • 23. Confer Roderick J. McIntosh’s otherwise excellent Peoples of the Middle Niger: The Island of Gold (Malden, MA: Blackwell, 1998), which barely mentions comparative Mande language study; and Raimund Kastenholz, Sprachgeschichte im West-Mande: Methoden und Rekonstruktionen (Cologne: Rüdiger Köppe, 1996).

  • 24. Bostoen, et al., “Early Bantu Expansion”; Koen Bostoen, Rebecca Grollemund, and Joseph Koni Muluwa, “Climate-Induced Vegetation Dynamics and the Bantu Expansion,” Comptes Rendus Geoscience 345 (2013): 336–349; Thomas E. Currie et al., “Cultural Phylogeography of the Bantu Languages of Sub-Saharan Africa,” Proceedings of the Royal Society: Biological Sciences 280, 1762 (2013): 1–8; Grollemund et al., “Bantu Expansion”; Christopher Ehret, “Bantu History: Big Advance, Although with a Chronological Contradiction,” Proceedings of the National Academy of Sciences 112, no. 44 (2015): 13428–13429.

  • 25. Bostoen et al., “Climate-Induced Vegetation Dynamics,” 339–345.

  • 26. Bostoen et al., “Climate-Induced Vegetation Dynamics,” 339–342; see also Kathryn de Luna “Conceptualizing Vegetation in the Bantu Expansion: Reflections on Linguistics in Central African History,” Quaternary International, 446 (2016): 158–168.

  • 27. Grollemund et al., “Bantu Expansion.”

  • 28. Stefanie Kahlheber, Koen Bostoen, and Katharina Neumann, “Early Plant Cultivation in the Central African Rainforest: First Millennium BC Pearly Millet from Southern Cameroon,” Journal of African Archaeology 7, no. 2 (2009): 253–272, here 260.

  • 29. Bernd Heine, H. Hoff, and Rainer Vossen, “Neuere Ergibnisse zur Territorialgeschichte der Bantu,” in Zur Sprachgeschichte und Ethnohistorie in Afrika, eds., Wilhelm Möhlig, Franz Rottland, and Bernd Heine (Berlin: Dietrich Reimer, 1977), 57–72; Christopher Ehret, “Bantu Expansions: Re-envisioning a Central Problem of Early African History,” International Journal of African Historical Studies 34, no. 1 (2001): 5–41; and Kairn Klieman, “The Pygmies Were Our Compass”: Bantu and Batwa in the History of West Central Africa, Early Times to c. 1900 (Portsmouth, NH: Heinemann, 2003).

  • 30. Confer Kathryn de Luna, Collecting Food, Cultivating People: Subsistence and Society in Central Africa (New Haven, CT: Yale University Press, 2016); Rhonda Gonzales, Societies, Religion, and History: Central-East Tanzanians and the World They Created, c. 200 BCE to 1800 (New York: Columbia University Press, 2009); Klieman, “‘Pygmies Were Our Compass’”; Christine Saidi, Women’s Authority and Society in Early East-Central Africa (Rochester: University of Rochester Press, 2010); David L. Schoenbrun, A Green Place, A Good Place: Agrarian Change, Gender, and Social Identity in the Great Lakes Region to the 15th Century (Portsmouth, NH: Heinemann, 1998); Seligman; “Determining Value”; and Rhiannon Stephens, A History of African Motherhood: The Case of Uganda, 700–1900 (New York: Cambridge University Press, 2013); and Vansina, How Societies Are Born.

  • 31. Stephens, African Motherhood, 187; Neil Kodesh, Beyond the Royal Gaze: Clanship and Public Healing in Buganda (Charlottesville: University of Virginia Press, 2010), 197; Schoenbrun, “Pythons Worked: Constellating Communities of Practice with Conceptual Metaphor in Northern Lake Victoria, ca. A.D. 800 to 1200,” in Knowledge in Motion: Constellations of Learning Across Time and Place, eds. Andrew P. Roddick and Ann B. Stahl (Tucson: University of Arizona Press, 2016): 216–246.

  • 32. John Roscoe, “Notes on the Manners and Customs of the Baganda,” Man: Journal of the Royal Anthropological Institute 31 (1901): 117–130, here 118–119; John Roscoe, The Baganda: An Account of Their Native Customs and Beliefs (London: Macmillan, 1911), 133; Sir Apolo Kagwa Kalibala Gulemye, Ekitabo kye Bika bya Baganda (Kampala: Uganda Bookshop, 1949), passim; and David Cohen, Historical Tradition of Busoga: Mukama and Kintu (Oxford: Clarendon Press, 1972), 6–7.

  • 33. Stephens, Motherhood, 24–25.

  • 34. Compare David Lee Schoenbrun, The Historical Reconstruction of Great Lakes Bantu Cultural Vocabulary: Etymologies and Distributions (Köln: Rüdiger Köppe, 1997), 89, to Stephens, Motherhood, 187.

  • 35. de Luna, “Classifying Botatwe”; Kathryn de Luna, “Hunting Reputations: Talent, Individuals, and Community in Precolonial South Central Africa,” Journal of African History 53, no. 3 (2012): 279–299; see also Vansina, How Societies Are Born, for different trajectories, to the northwest of de Luna’s region of interest.

  • 36. de Luna, Collecting Food, Cultivating People, 126.

  • 37. de Luna, Collecting Food, Cultivating People, 205–206.

  • 38. de Luna, Collecting Food, Cultivating People, 173.

  • 39. Jeff Marck and Koen Bostoen, “Proto-Oceanic (Austronesian) and Proto-East Bantu Society (Niger Congo) Residence, Descent, and Kin Terms, ca. 1000 BC,” in Kinship, Language and Prehistory: Per Hage and the Renaissance in Kinship Studies, eds. Doug Jones and Bojka Milicic (Salt Lake City: University of Utah Press, 2011), 83–91; and Robert S. Walker and Marcus J. Hamilton, “Social Complexity and Linguistic Diversity in the Austronesian and Bantu Population Expansions,” Proceedings of the Royal Society B: Biological Sciences 278, no. 1710 (2010): 1399–1404.

  • 40. de Luna, Collecting Food, Cultivating People, 172–219, explores additional examples. For different trajectories in the history of trade and power to the south of de Luna’s region, see Carolyn Hamilton, Bernard Mbenga, and Robert Ross, eds., Cambridge History of South Africa, vol. 1 (Cambridge, UK: Cambridge University Press, 2009); James Denbow, Carla Klehm, and Laure Dussubieux, “The Glass Beads of Kaitshàa and Early Indian Ocean Trade into the Far Interior of Southern Africa,” Antiquity 89, no. 344 (2015), 361–377; Shadreck Chirikure, “Land and Sea Links: 1500 Years of Connectivity between Southern Africa and the Indian Ocean Rim Regions, ad 700 to 1700,” African Archaeological Review 31, no. 4 (2014), 705–724; Paul Landau, Popular Politics in the History of South Africa, 1400–1948 (New York: Cambridge University Press, 2010), 42–73; and Raemin Jimenez, “Rites of Reproduction: Tradition, Political Ethics, Gender, and Generation among Nguni-speakers of Southern Africa, 8th–19th Centuries CE,” Unpublished PhD dissertation, Northwestern University, 2017.

  • 41. de Luna, Collecting Food, Cultivating People, 196–200 (Njila); 202–209 (Sabi, Luba); 56ff (Kusi). For different trajectories north of de Luna’s region, see Pierre de Maret, “From Kinship to Kingship: An African Journey into Complexity,” Azania 47, no. 3 (2012): 314–326; Pierre de Maret, “The Power of Symbols and the Symbols of Power through Time: Probing the Luba Past,” in Beyond Chiefdoms: Pathways to Complexity in Africa, ed. Susan Keech McIntosh (Cambridge, UK: Cambridge University Press, 1999), 151–165.

  • 42. de Luna, Collecting Food, Cultivating People, 213. For related trajectories to the east and southeast of de Luna’s region, see Susan Keech McIntosh and Brian M. Fagan, “Re-dating the Ingombe Ilede Burials,” Antiquity 91, no. 358 (August 2017), 1069–1077; Innocent Pikirayi, The Zimbabwe Culture: Origins and Decline of Southern Zambezian States (Walnut Creek, CA: AltaMira, 2001); and Shadreck Chirikure et al., “New Pathways of Sociopolitical Complexity in Southern Africa,” African Archaeological Review 30, no. 4 (2013), 339–366.

  • 43. Schoenbrun, “Pythons Worked,” 221, revising Schoenbrun, A Green Place, 200.

  • 44. Koen Bostoen, “Semantic Vagueness and Cross-Linguistic Lexical Fragmentation in Bantu: Impeding Factors for Linguistic Palaeontology,” Sprache und Geschichte in Afrika 20 (2009): 51–64.

  • 45. William Hanks, Language and Communicative Practices (Boulder, CO: Westview, 1996), 239ff.

  • 46. In West Africa, Moustapha Sall, “Academic Research in West Africa: The Case of Senegal,” in Field Manual for African Archaeology, eds. Alexandre Livingstone Smith, Els Cornelissen, Olivier P. Gosselain, and Scott MacEachern (Tervuren, Belgium: Royal Museum for Central Africa, 2017), 18–23; Abidemi Babatunde Babalola et al., “Ile-Ife and Igbo Olokun in the History of Glass in West Africa,” Antiquity 91, no. 357 (2017): 732–759; Sam Nixon, “Excavating Essouk-Tadmakka (Mali): New Archaeological Investigations of Early Islamic Trans-Saharan Trade,” Azania 44, no. 2 (2009): 217–255; Timoth Insoll et al., “Insights into Past Ritual Practive at Yikpabonogo, Northern Region, Ghana,” African Archaeological Review 30, no. 4 (2013): 475–499; Wazi Apoh and Kodzo Gavua, “Material Culture and Indigenous Spiritism: The Katamansu Archaeological ‘Otutu’ (Shrine),” African Archaeological Review 27, no. 3 (2010) 211–235; and William Narteh Gblerkpor, “Archaeological Indications of Past Lifeways on Krobo Mountain, Ghana,” Ghana Social Science Journal 5/6, nos. 1/2 (2009): 153–187. In Southern Africa and beyond, Innocent Pikirayi, “The Archaeology of sub-Saharan Africa: An Overview,” in The Oxford Handbook of Archaeology, eds. Barry Cunliffe, Chris Gosden, and Rosemary A. Joyce (Oxford: Oxford University Press, 2009), 723–762; Peter Mitchell, The Archaeology of Southern Africa (Cambridge, UK: Cambridge University Press, 2002); Innocent Pikirayi and Shadreck Chirikure, “Debating Great Zimbabwe,” Azania 46, no. 2 (2011): 221–231. In Northeastern Africa, see David W. Phillipson, Foundations of an African Civilisation: Aksum and the northern Horn, 1000 BCAD 1300 (Oxford: James Currey, 2012).

  • 47. Over time, of course, human society has followed this trajectory; see “States of Complexity,” S[anta] F[e] I[nstitute] Bulletin 27 (2013); but for different reasons and with different social arrangements. For example, social complexity developed outside of political centralization in a number of places; see McIntosh, Peoples of the Middle Niger; Susan Keech McIntosh, “Pathways to Complexity: An African Perspective,” in Beyond Chiefdoms, ed., Susan Keech McIntosh (Cambridge, UK: Cambridge University Press, 1999), 1–30; David W. Phillipson, Ancient Ethiopia: Aksum, Its Antecedents and Successors (London: British Museum Press, 1998); Akinwumi Ogundiran, “Four Millennia of Cultural History in Nigeria (ca. 2000 B.C.–A.D. 1900): Archaeological Perspectives,” Journal of World Prehistory 19 (2005): 133–168; Vansina, Paths; and Ann Brower Stahl, “Political Economic Mosaics: Archaeology of the Last Two Millennia in Tropical Sub-Saharan Africa,” Annual Review of Anthropology 33 (2011): 145–172.

  • 48. Calling such work “ethnography” implies a limit imposed by the specifics of the ethnic, corporate, or other group to which such work applies, while calling guild historical writing about past cultures “cultural history” might suggest no similar limit. The distinction is misleading, not only because “cultural history” always focuses on some particular cultural field, but also because both kinds of writing draw on conventions of genre and logics of induction and deduction. Just because they do not always draw on the same conventions does not mean that one is less partial than the other. See discussion in the section “Traditions, Written and Oral.”

  • 49. For a careful explanation of this and other methods of using “ethnographic” evidence with archaeology—from identifying core research problems to interpreting excavated and survey information—see Ann B. Stahl, Making History in Banda: Anthropological Visions of Africa’s Past (Cambridge, UK: Cambridge University Press, 2001), chapters 1 and 2. For a broader set of reflections on the analogical processes at work in such method, see Alison Wylie, Thinking from Things: Essays in the Philosophy of Archaeology (Berkeley: University of California Press, 2002); for the history of such work in Africa, see Scott MacEachern, “Foreign Countries: The Development of Ethnoarchaeology in Sub-Saharan Africa,” Journal of World Prehistory 10, no. 3 (1996), 243–304; Diane Lyons, “Ethnoarchaeology,” in The Oxford Handbook of African Archaeology, eds. Peter Mitchell and Paul J. Lane (Oxford: Oxford University Press), 87–101; for an important demonstration of its value in revising archaeological conventions, see Shadreck Chirikure, Munyaradzi Manyanga, and A. Mark Pollard, “When Science Alone Is Not Enough: Radiocarbon, Timescales, History, Ethnography, and Elite Settlements in Southern Africa,” Journal of Social Archaeology 12, no. 3 (2012): 356–379.

  • 50. Ann Brower Stahl, “Colonial Entanglements and the Practices of Taste: An Alternative to Logocentric Approaches,” American Anthropologist 104, 3 (2002): 827–845. But see Wendy Ashmore, Archaeology of a Continent, in Theory,” in Stephanie Wynne-Jones and Jeffrey B. Fleisher, eds. Theory in Africa, African in Theory: Locating Meaning in Archaeology (New York: Routledge, 2015), 271–279; Zoe Crossland, Encounters with Ancestors in Highland Madagascar: Material Signs and Traces of the Dead (New York: Cambridge University Press, 2014), Chapters 1 and 2; Scott Ortman, Winds from the North: Tewa Origins and Historical Anthropology (Salt Lake City: University of Utah Press, 2012), Chapter 9; the same problems hold for ancient graphic systems no longer practiced, see Stephen D. Houston, “The Archaeology of Communication Technologies,” Annual Review of Anthropology 33 (2004), 223–250.

  • 51. Peter T. Robertshaw, A History of African Archaeology, Portsmouth, NH: Heinemann, 1990; Ann Brower Stahl, ed., African Archaeology: A Critical Introduction (Oxford: Blackwell, 2005); Peter Mitchell and Paul J. Lane, eds., The Oxford Handbook of African Archaeology (Oxford: Oxford University Press, 2013); for an open-access guide to the practice of archaeology in Africa see Livingstone Smith et al., Field Manual for African Archaeology. See also, Stephanie Wynne-Jones and Jeffrey D. Fleisher, eds., Theory in Africa; Africa in Theory: Locating Meaning in Archaeology (New York: Routledge Publishers, 2016). For an excellent example of the best of all these approaches, see Akinwumi Ogundiran, “Of Small Things Remembered: Beads, Cowries, and Cultural Translations of the Atlantic Experience in Yorubaland,” International Journal of African Historical Studies 35, no. 2–3 (2002): 427–457. Art historians have taken up similar moves. Less explicitly historical, see Rowland Abiodun, Yoruba Art and Language: Seeking the African in African Art (New York: Cambridge University Press, 2014); set in a particular historical “moment,” see Suzanne Preston Blier, Art and Risk in Ancient Yoruba: Ife History, Power, and Identity, c. 1300 (New York: Cambridge University Press, 2014). Rock art study is particularly rich; see David Lewis Williams, The Mind in the Cave: Consciousness and the Origins of Art (London: Thames & Hudson, 2002); Augustin Holl, Saharan Rock Art: Archaeology of Tassilian Pastoralist Iconography (Walnut Creek, CA: AltaMira Press, 2004); and Catherine Namono, Catherine. “Resolving the Authorship of the Geometric Rock Art of Uganda,” Journal of African Archaeology 8 (2010): 239–257.

  • 52. The scholar-traveler Ibn Battuta’s writings are famous in this genre; see Ross E. Dunn, The Adventures of Ibn Battuta: A Muslim Traveler of the Fourteenth Century, 3rd ed. (Berkeley: University of California Press, 2012).

  • 53. Stephanie Wynne-Jones, A Material Culture: Consumption and Materiality on the Coast of Precolonial East Africa (Oxford: Oxford University Press, 2016).

  • 54. Jeffrey B. Fleisher et al., “When Did the Swahili Become Maritime?” American Anthropologist 117, no. 1 (2015): 100–115; and Stephanie Wynne-Jones and Adria Laviolette, eds., The Swahili World (New York: Routledge, 2017).

  • 55. Derek Nurse and Thomas Hinnebusch, Swahili and Sabaki: A Linguistic History (Berkeley: University of California Press, 1993).

  • 56. Wynne-Jones and Laviolette, The Swahili World. The well-established moniker misleads—thus, the scare quotes—because people speaking other Bantu languages and languages other than Bantu have lived and now live along and just behind this coastline; cf. Nurse and Hinnebusch, Swahili and Sabaki.

  • 57. Wynne-Jones, Material Culture, 144; and Nurse and Hinnebusch, Swahili and Sabaki.

  • 58. Nurse and Hinnebusch, Swahili and Sabaki, 19–23, with caveats and explanations; Christopher Ehret, An African Classical Age, 184–189.

  • 59. Wynne-Jones, Material Culture, 44–54.

  • 60. Marilee Wood, “East Africa and the Indian Ocean World in the First Millennium CE: The Glass Bead Evidence,” in Early Exchange Between Africa and the Wider Indian Ocean World, Gwynn Campbell, ed., (New York: Palgrave Macmillan, 2016), 173–193; and Bing Zhao, “Chinese-style ceramics in East Africa from the 9th to the 16th century: A case of changing value and symbols in the multi-partner global trade,” Afriques: Debats, methods et terrains d’histoire 6 (2015).

  • 61. Fleisher et al., “When Did the Swahili Become Maritime?” 107–110.

  • 62. Jeffrey Fleisher, “Rituals of Consumption and the Politics of Feasting on the Eastern African Coast, AD 700–1500,” Journal of World Prehistory 23, no. 4 (2010): 195–217.

  • 63. Stephanie Wynne-Jones, “Remembering and Reworking the Swahili Diwanate: The Role of Objects and Places at Vumba Kuu,” International Journal of African Historical Studies 43, no. 3 (2010): 407–427.

  • 64. Jeffrey Fleisher, “Situating the Swahili House,” in Wynne-Jones and Fleisher, Theory in Africa, 72–89.

  • 65. See the 3D digital reconstructions available at the website Songo Mnara: Urban Space, Social Memory, and Materiality of Jeffrey Fleisher, one of site’s principal investigators.

  • 66. Wynne-Jones, Material Culture, 165–166.

  • 67. Ceri Ashley, “Toward a Socialised Archaeology of Ceramics in Great Lakes Africa,” African Archaeological Review 27, no. 2 (2010), 135–163; Andrew Reid, “Buganda: Unearthing an African Kingdom” Archaeology International 7 (2003): 40–43; and Andrew Reid and Ceri Z. Ashley, “Islands of Agriculture on Victoria Nyanza,” in Archaeology of African Plant Use, eds. Chris J. Stevens, Sam Nixon, Mary Anne Murray, and Dorian Q. Fuller (Walnut Creek, CA: Left Coast Press, 2014), 179–188.

  • 68. Andrew Reid and Ceri Z. Ashley, “A Context for the Luzira Head,” Antiquity 82, no. 315 (2008): 99–112.

  • 69. Jan Vansina, Oral Tradition as History (Madison: University of Wisconsin Press, 1985), 27.

  • 70. Bethwell Alan Ogot, “The Construction of Luo Identity and History,” in African Words, African Voices: Critical Practices in Oral History, eds. Luise White, Stephan Miescher, and David W. Cohen (Bloomington: Indiana University Press, 2001), 31–52.

  • 71. Oral traditions assembled by African intellectuals in West African contexts include Samuel Johnson, The History of the Yorubas: From the Earliest Times to the Beginning of the British Protectorate (Lagos: Church Missionary Society Bookshops, 1921) and Jacob U. Egharevba, A Short History of Benin, 3rd ed., revised (Ibadan: Ibandan University Press, 1960). See further citations in John K. Thornton, “European Documents,” in Philips, Writing African History.

  • 72. Christopher Wrigley, Kingship and State: The Buganda Dynasty (Cambridge, UK: Cambridge University Press, 1996), 28–34.

  • 73. Bethwell Alan Ogot, History of the Southern Luo, vol. 1 (Oxford: Oxford University Press, 1967); and Jan Bender Shetler, Imagining Serengeti: A History of Landscape Memory in Tanzania from Earliest Times to the Present (Athens: Ohio University Press, 2007), 18–25.

  • 74. David Schoenbrun, “A Mask of Calm: Emotion and Founding the Kingdom of Bunyoro in the Sixteenth Century,” Comparative Studies in Society and History 55, no. 3 (2013): 634–664, here 645–664.

  • 75. David William Cohen, “The Undefining of Oral Tradition,” Ethnohistory 36, no. 1 (1989): 9–18; and Kodesh, Beyond the Royal Gaze, 1–7, 20–26.

  • 76. John Dillery, Clio’s Other Sons: Berossus and Manetho (Ann Arbor: University of Michigan Press, 2015), 121.

  • 77. Holger Bernt Hansen, “The Colonial Control of Spirit Cults in Uganda,” in Revealing Prophets, eds. David M. Anderson and Douglas H. Johnson (Oxford: James Currey, 1995), 143–163; Steven Feierman, “Colonizers, Scholars, and the Creation of Invisible Histories,” in Beyond the Cultural Turn, eds. Victoria E. Bonnell and Lynn Hunt (Los Angeles: University of California Press, 1999), 181–216.

  • 78. For an example from West Africa, see Carola Lentz, Land, Mobility, and Belonging in West Africa (Bloomington: Indiana University Press, 2013).

  • 79. Kagwa, Bika, 38.

  • 80. Schoenbrun, “Pythons Worked.”

  • 81. The intellectual J. T. K. Ggomotoka, Head of the Princes in the oligarchy from 1919–1941, used it to make analogies with European political charters, such as touching on Clovis, in arguing that Kintu came from outside Buganda; see J. T. K. Ggomotoka, “Ebye Buganda: Ekika ky’Abalangira b’omu Buganda,” Munno 10, no. 115 (1920): 121–122; 131–133 (analogies); and J. T. K. Ggomotoka, “Ebye Buganda: Sekabaka Kintu nga bwe yajja,” Munno 10, no. 116 (1920): 150–152.

  • 82. Antoine Meillet, “Comment les mots changent de sens,” Année Sociologique 9 (1906): 1–38; and Andreas Blank, “Words and Concepts in Time: Towards Diachronic Cognitive Onomasiology,” in Words in Time, ed. R. Eckardt (Berlin: de Gruyter, 2003), 37–65. Metaphor and conceptual blending (explored below) are other such agents of making meaning.

  • 83. Scott Ortman, “Conceptual Metaphor in the Archaeological Record: Methods and an Example from the American Southwest,” American Antiquity 65, no. 4 (2000): 613–645; Scott Ortman, “Bowls to Gardens: A History of Tewa Community Metaphors,” in Religious Transformation in the Late Pre-Hispanic Pueblo World, eds. Donna M. Glowacki and Scott Van Keuren (Tucson: University of Arizona Press, 2013), 84–108; David Schoenbrun, “Pythons Worked”; and Axel Fleisch and Rhiannon Stephens, eds., Doing Conceptual History in Africa, (New York: Berghahn Books, 2016), use the conceptual to do this.

  • 84. William Shakespeare, Hamlet, Act III, Scene I, line 1751.

  • 85. Webb Keane, “Semiotics and the Social Analysis of Material Things,” Language and Communication 23 (2003): 409–425, here, 409, 419ff.

  • 86. For the logic of a cuneiform system of knowledge that dwells on observed reality to explain the meaning of things rather than to explain how structures in the world worked, see Francesca Rochberg, Before Nature: Cuneiform Knowledge and the History of Science (Chicago: University of Chicago Press, 2016); and for a famous example set in Africa, see Edward E. Evans-Pritchard, Witchcraft, Oracles, and Magic Among the Azande (London: Faber & Faber, 1937).

  • 87. Zoltán Kövecses, Metaphor in Culture: Universality and Variation (Cambridge, UK: Cambridge University Press, 2005); Ortman, Winds from the North, 203–225; and Schoenbrun, “Mixing, Moving, Making, Meaning,” 297–300.

  • 88. Kövecses, Metaphor in Culture, 68.

  • 89. Ortman, “Bowls to Gardens,” 89–96; and Schoenbrun, “Pythons Worked,” 231–239.

  • 90. Joseph Grady et al, “Blending and Metaphor,” in Metaphor in Cognitive Linguistics, ed. R. Gibbs and G. J. Steen (Amsterdam: John Benjamins, 1999), 101–124; and Ortman, “Conceptual Metaphor,” 618.

  • 91. Edwin Hutchins, “Material Anchors for Conceptual Blends,” Journal of Pragmatics 37 (2005): 1555–1577, 1563 (quoted material). This article explores other, far more complicated blends, such as those involved in Micronesian navigation techniques.

  • 92. Reid and Ashley, “A Context for the Luzira Head,” 104–108 (dating); Schoenbrun, “Pythons Worked.”

  • 93. Developing new ideas from established ones makes it easier for the new idea to “stick,” see Karin Barber, “Improvisation and the Art of Making Things Stick,” in Creativity and Cultural Improvisation, ed. Elizabeth Hallam and Tim Ingold (Oxford: Berg, 2007), 25–41.

  • 94. Roscoe, Baganda, 327–328.

  • 95. Eve Sweetser, From Etymology to Pragmatics: Metaphorical and Cultural Aspects of Semantic Structure (Cambridge, UK: Cambridge University Press, 1990).

  • 96. Jonathon Glassman, “Ethnic and Racial Thought in Africa,” in Handbook of Modern African History, ed., William Worger (New York: Wiley-Blackwell, 2018).

  • 97. A fine example, though not set out in historical terms, is Michael Schatzberg, Political Legitimacy in Middle Africa: Father, Family, Food (Bloomington: Indiana University Press, 2001).

  • 98. Carolyn Hamilton, Terrific Majesty: The Powers of Shaka Zulu and the Limits of Historical Invention (Cambridge, MA: Harvard University Press and David Philip, 1998), Introduction; Naomi Quinn, “The Cultural Basis of Metaphor,” in Beyond Metaphor, ed., James Fernandez (Stanford, CA: Stanford University Press, 1991), 56–93; and Webb Keane, “Marked, Absent, Habitual: Approaches to Neolithic Religion at Catalhöyük,” in Religion in the Emergence of Civilization: Catalhöyük as a Case Study, ed., Ian Hodder (Cambridge, UK: Cambridge University Press, 2010), 187–219, here, 193, fn3.

  • 99. Thomas J. Csordas, “Embodiment as a Paradigm for Anthropology,” Ethos 18, no. 1 (1990): 5–47; Jordan Zlatev, Situated Embodiment: Studies in the Emergence of Spatial Meaning (Stockholm: Gotab Press, 1997); and George Lakoff and Mark Johnson, Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Thought (New York: Basic Books, 1999), 508–509;

  • 100. Jane I. Guyer, “Africa Has Never Been ‘Traditional’,” African Studies Review 50, no. 2 (2007): 183–202; and Gareth Austin, “Reciprocal Comparison and African History: Tackling Conceptual Eurocentrism in the Study of Africa’s Economic Past,” African Studies Review 50, no. 3 (2007): 1–28.

  • 101. Edward Sapir, Time Perspective in Aboriginal American Culture: A Study in Method, Department of Mines, Geological Survey Memoir 90, Anthropological Series 13 (Ottawa: Government Printing Bureau, 1916).

  • 102. Jan Vansina, Raymond Mauny, and L. V. Thomas (eds.), The Historian in Tropical Africa (Oxford: Oxford University Press, 1964); Daniel F. McCall, Africa in Time-Perspective: A Discussion on Historical Reconstruction from Unwritten Sources (Boston and Legon: Boston University Press and Ghana University Press, 1964); Ogot, History of Southern Luo; Creighton Gabel and Norman R. Bennett, eds., Reconstructing African Culture History (Boston: Boston University Press, 1967); Roger Blench and Matthew Spriggs, eds., Archaeology and Language, 4 vols. (London: Routledge Publishers, 1997–1999); Christopher Ehret, Southern Nilotic History (Evanston, IL: Northwestern University Press, 1971); Ehret, African Classical Age; Ehret, History and Testimony of Language; Jan Vansina, The Children of Woot: A History of the Kuba Peoples (Madison: University of Wisconsin Press, 1978); Vansina, Paths; Vansina, How Societies Are Born; Thomas Q. Reefe, The Rainbow and the Kings: A History of the Luba Empire to 1891 (Los Angeles: University of California Press, 1981); Peter Schmidt, Historical Archaeology: A Structural Approach in an African Society (Westport, CT: Greenwood Press, 1978); Peter Schmidt, Iron Technology in East Africa: Symbolism, Science, and Archaeology (Bloomington, IN: Indiana University Press, 1997); Schoenbrun, A Green Place; Schoenbrun, “Pythons Worked”; Klieman, ‘Pygmies Were Our Compass’; Gonzales, Society, Economy, and Religion; Landau, Popular Politics; Roger Blench, Archaeology, Language, and the African Past (Walnut Creek, CA: AltaMira Press, 2006); Saidi, Women’s Authority; Kodesh, Beyond the Royal Gaze; Ann Haour, Outsiders and Strangers: An Archaeology of Liminality in West Africa (Oxford: Oxford University Press, 2013); Shetler, Imagining Serengeti; Amanda L. Logan, “‘Why Can’t People Feed Themselves?’: Archaeology as Alternative Archive of Food Security in Banda, Ghana,” American Anthropologist 118, no. 3 (2016): 508–524; Edda Fields-Black, Deep Roots: Rice Farmers in West Africa and the African Diaspora (Bloomington, IN: Indiana University Press, 2008); Stephens, Motherhood; de Luna, Collecting Food.

  • 103. Ian Morris, Archaeology as Cultural History: Words and Things in Ancient Greece (Malden, MA: Blackwell Publishers, 2000); Patrick V. Kirch and Roger C. Green, Hawaiki, Ancestral Polynesia: An Essay in Historical Anthropology (New York: Cambridge University Press: 2001); and Ortman, Winds from the North.

  • 104. For a good example of the new economic history, see Alain Bresson, The Making of the Ancient Greek Economy: Institutions, Markets, and Growth in the City-States, trans. Steven Rendall (Princeton, NJ: Princeton University Press, 2016). Atlantic Africa has been well served by Jane I. Guyer, Marginal Gains (Chicago: University of Chicago Press, 2004).

  • 105. John D. Giblin, “A Reconsideration of Rwandan Archaeological Ceramics and their Political Significance in a Post-Genocide Era,” African Archaeological Review 30, no. 4 (2013), 501–529; Jan Vansina, Antecedents of Modern Rwanda: The Nyiginya Dynasty (Madison: University of Wisconsin Press, 2004); and Carolyn Hamilton, John Wright, and Bernard Mbenga, eds., Cambridge History of South Africa, vol. 1, “Introduction” (Cambridge, UK: Cambridge University Press, 2010).

  • 106. Paul Landau, Popular Politics and the History of South Africa, 1400–1948 (New York: Cambridge University Press, 2010); Carolyn Hamilton and John Wright, “Moving Beyond Ethnic Framing: Political Differentiation in the Chiefdoms of the KwaZulu-Natal Region before 1830,” Journal of Southern African Studies 43, 4 (2017), 663–679; and Glassman, “Ethnic and Racial Thought in Africa,” passim; Achille Mbembe, Critique de la raison nègre (Paris: Éditions de la Découverte, 2013).

  • 107. Confer Yvonne Bastin, et al., Bantu Lexical Reconstructions III and the materials available at KONGOKING: An Interdisciplinary Approach to the Origins of the Kongo Kingdom[ based at Gent University, Belgium.

  • 108. Richard J. Reid, “Ghosts in the Academy: Historians and Historical Consciousness in the Making of Modern Uganda,” Comparative Studies in Society and History 56, no. 2 (2015): 351–380.

  • 109. CODESRIA (Council for the Development of Social Science Research in Africa), has a Small Grants Program and an Advanced Research Fellowship Program; OSSREA (Organization for Social Science Research in Eastern and Southern Africa) organizes Special Projects, with outside funding; The ACLS (American Council of Learned Societies) has funding programs in the African Humanities, for dissertation completion and postdoctoral research and writing for scholars based in Africa with “a current affiliation at an institution in Ghana, Nigeria, Tanzania, or Uganda.

  • 110. For pollen studies, see Observatoire Midi-Pyrénées. For historical linguistics, see Christopher Ehret, A Historical-Comparative Reconstruction of Nilo-Saharan (Cologne: Rüdiger Köppe Press, 2001); Christopher Ehret, Reconstructing Proto-Afroasiatic (Proto-Afrasian): Vowels, Tone, Consonants, and Vocabulary (Berkeley: University of California Press, 1995); and for Bantu, see Yvonne Bastin, André Coupez, Evariste Mumba, and Thilo C. Schadeberg, Bantu Lexical Reconstructions3. As a partial guide to the latter, see Koen Bostoen and Yvonne Bastin, “Bantu Lexical Reconstruction, Oxford Handbooks Online. On lexicostatistics, see George Starostin and Phil Krylov, The Global Lexicostatistics Database. Datasets can also be found at the Archaeology Data Service website. For oral traditions, see the International Society for the Study of Oral Traditions.