Is it possible or useful to write a truly comparative history of race? The question may seem frivolous, given the existence of countless excellent comparative studies. But most of those studies focus on various iterations of white supremacy, or on its inflections in Pan-Africanism or other forms of anti-racist racial modernism.Footnote ¹ What of other forms of racial thought, forms that have nothing to do with ideologies of white supremacy or are tied to it only indirectly? In the literature on Africa (which will be the focus of this essay), the most obvious examples are those that shaped violence between Hutu and Tutsi in Rwanda, or between so-called Arabs and Africans in Darfur or Zanzibar. There are other instances, as we will see. They all raise historical questions of how locally-specific ways of thinking about difference might become racialized—that is, how they might become invested with explicit meanings of bodily descent or of “blood,” to use a ubiquitous metaphor.

When dealing with race or any other kind of ethnicity, historians of Africa must contend with two opposing tendencies. The first are assumptions of primordial ethnic essences that still cling to popular perceptions of Africa. This was readily apparent in the press coverage of the 1994 Rwandan genocide, when journalists lazily wrote about ancient tribal hatreds that erupted in the absence of a strong colonial state. Primordialism of course is part of the basic myth of the African past, in which Africans appear as inherently “tribal” beings, congenitally incapable of transcending their inherited ethnic essence: as Hugh Trevor-Roper put it in an infamous reiteration of Hegel, Africa's past consisted not of history, but merely of the “unrewarding gyrations of barbarous tribes.”Footnote ² As historians labored to refute such myths throughout the last third of the twentieth century, one of their recurring themes was the modern origins of ethnic thought. But in a sense, they were too successful. Since the 1990s, accounts of African ethnicity have hewed to a resolute modernism just as often as to old-school primordialism: journalists and scholars nowadays commonly write that ethnic differences were invented by the twentieth-century colonial state.Footnote ³

In recent decades, this modernism has been revised by scholars who emphasize what are sometimes called “constructivist” factors, after a parallel literature in the study of nationalism. Without discounting the impact of colonialism, these authors have demonstrated how African intellectuals actively crafted new forms of ethnic thought out of indigenous cultural materials. Modern tribalism did not just spring into existence in response to colonial influences; rather, it built on locally inherited discourses of belonging and difference.Footnote ⁴ But although such constructivist revisions have profoundly deepened our historical perspectives on so-called tribalism, they have not had as much impact in studies of the forms of ethnic thought that are often described as “race.” In that literature, a nagging modernism persists, and it is instructive to consider why.

To do so, however, we must specify what we mean by “race.” I have mentioned the central idea of bodily descent, or “blood.” Yet, in this regard it is crucial to recognize that no hard and fast analytic distinction can be made between “race” and other forms of ethnicity. (Historical distinctions are a different matter, inasmuch as a historical approach demands that we recognize each instance of ethnic or racial thought, or of any phenomenon, as unique. That is why we cannot pretend that American white supremacy, for example, is somehow identical to other instances of race or ethnicity: to say that two historical phenomena are comparable is not to suggest that they are commensurate.Footnote ⁵) Forms of thought we call “ethnic,” like those we call “race,” can all be characterized as categorical orders that distinguish differences between “human kinds” through metaphors of common descent.Footnote ⁶

An explication of these two terms will clarify matters. My use of the Foucauldian notion of categorical order, which I take from Liisa Malkki's writings on ethnic nationalism, is akin to Ann Stoler's concept of a racial “regime of truth,” or Rogers Brubaker's argument that race and ethnicity are matters of cognition, perspectives “on the world” before they are things “in the world.”Footnote ⁷ I use “race,” in other words, as shorthand for racial thought, the modes of perception by which people read the social and/or somatic clues that denote racial categories. These modes of categorization are historical creations, each specific to a particular society; as scholars have long observed, they do not arise from naturally occurring distinctions but, in many ways, create distinctions.Footnote ⁸ For purposes of historical analysis, it is useful to distinguish these habits of thought from the explicit ideas and practices of exclusion or domination that we call racism. For those socialized in American society, for example, the history of racism is obviously an important part of our history of racial thought. But the two are not necessarily identical. Hence, in societies structured by systematic racism, members of the subordinate stratum may share prevailing perceptions of racial difference, and actively use those perceptions to craft strategies and philosophies of liberation, without partaking in the convictions (and certainly not the practices) of innate superiority and inferiority properly described as racist.Footnote ⁹

Scholars since Weber have observed that virtually all notions of ethnic or national thought revolve around metaphors of descent.Footnote ¹⁰ Such metaphors can be taken more explicitly, or less. At one end of the spectrum are vague notions of common ancestry, used as little more than figures of speech, as when American politicians speak of the “Founding Fathers.” At the other end are discourses that perceive common descent not simply as metaphor but as something real and significant, fixed in “the blood” or, to use the currently fashionable scientific metaphor, the genes.Footnote ¹¹ In the postwar world we usually describe only this second kind of thinking as “racial.” But the boundary between the latter form and forms in the middle of the spectrum—call them ethnicity, tribalism, or “culture”Footnote ¹²—is not absolute. An “aura of descent,” at the least, hovers over them all.Footnote ¹³ And history shows how readily any of them can become transformed: to take an immediate example, consider how birtherism during the Obama administration and debates over the Fourteenth Amendment's Citizenship Clause since 2016 have revived and reformulated many Americans’ convictions about the racial nature of their ostensibly civic nation.Footnote ¹⁴ Likewise, as we will see, tensions over immigration and neoliberalism have begun to racialize civic nationalisms in parts of Africa. Any attempt to pose a sharp distinction can only obscure the processes of racialization by which diverse modes of thinking about “human kinds” can become invested with more explicit ideas of descent.

Some scholars locate a categorical distinction between race and other forms of ethnicity in the criterion of hierarchy. According to this view, racial thought involves imagining ethnic categories as “horizontal” strata, linked to one another in relations of inequality that structure a single social formation. (A focus on the practices of domination that structure such “racist social formations” has produced some of the most thorough analyses of the sociological workings of racism in the modern West.Footnote ¹⁵) Discourses of Tutsi and Hutu difference fit this model well. It is one we can contrast with the “vertical” divisions that in African contexts are typically described as “tribes.” Whereas “races,” in this formulation, constitute “ranked” strata within a single society, tribalist thought imagines each ethnic unit as an “incipient whole society.”Footnote ¹⁶ (Anywhere other than Africa, “tribalism” would be described as “ethnic nationalism.”) It is this latter form, of “tribalism” or “unranked ethnicity,” that has received some of the most nuanced historical treatments in the Africa literature, treatments that avoid a limiting modernism without being primordialist.Footnote ¹⁷

To return to our question, then: Why such persistent modernism in Africanist studies of race—in studies, that is, of discourses of difference that emphasize explicit concepts of bodily descent and/or hierarchy? One reason stems from the common assumption that racial thought originated in the West. There is, in fact, an influential comparative and social science literature that defines race that way: as a mode of thought invented by Europe in the course of its imperial expansion.Footnote ¹⁸ (This view ignores the fact that many key ideas in Western racial thought, including those that were deployed to explain the inferiority of colonial subjects, were first elaborated to explain differences among Europeans themselves—between Gauls and Franks, for example—long before the Columbian voyages.Footnote ¹⁹) When racial thought is found in non-Western intellectual traditions, then, it is understood to have been introduced as part of the toolkit of empire and transmitted to the subject population.Footnote ²⁰ In most of Africa, that would have been after the 1890s. In Rwanda, for example, where Belgians ruled through the old monarchy, members of the Tutsi elite sent their sons to study at mission schools after World War One. There, they learned racial theories that explained why it was that some Africans had been able to build complex states and military structures such as the Rwandan kingdom. These aristocratic students learned that they were descended from an advanced race of “Hamites,” who long ago had migrated from the north and brought statecraft and other civilizing arts to the indigenous Hutu, aborigines of inferior “negroid” stock. The language of these “Hamitic” theories seems to confirm another modernist assumption common in the literature: that race originated as scientific doctrine.Footnote ²¹ Again, this assumption would imply that, if found outside the West, racial thought must have begun with colonialism.

Among the many problems with this view is its distortion of the history of race in the West itself. Biological doctrines, in fact, came late to Western racial thought, and their dominance was relatively fleeting, roughly from the mid-nineteenth century to World War Two.Footnote ²² Even in its heyday, race-science was hardly the only game in town; colonial racisms, for example, rarely emphasized biological difference.Footnote ²³ (The Belgian valorization of a “Hamitic” racial elite was far from the norm.) Western racial thought, in fact, grew from multiple sources, many of which had little to do with biology. One of the most significant in the modern era was stadial historicism: the cluster of ideas, often traced to the Enlightenment, that understands history as a progression through a set succession of stages.Footnote ²⁴ In this perspective, the problem of African difference (say) is explained not by reference to fixed biological qualities, but by Africans’ low position on the ladder of progress from savagery to civilization—so low, in fact, that few had ever embarked on the construction of stable civil orders before Europeans intervened. Colonial racisms, for the most part, were of this stadial, historicist kind, emphasizing social evolution more than biology, especially after World War One. Their central quality was a paternalist conviction that colonial rulers and educators were engaged in a civilizing mission to draw their subjects along the path of progress.

There is no doubt that Western concepts, historicist and, less frequently, biological, influenced how colonial subjects thought about difference. But the key intellectual work of translating those concepts into local terms—crafting narratives, for example, in which Hutu and Tutsi resembled “Gauls and Franks”Footnote ²⁵—was not performed by Europeans. Rather, it was performed by Africans themselves: schoolteachers especially, but also moral reformers, amateur historians, and other subaltern intellectuals. And the minds of those intellectuals were not blank slates. In Rwanda, they bore the imprint of the tumultuous century that preceded European conquest, which saw the violent expansion of the Nyiginya kingdom and attendant processes of acute polarization. The centralizing dynasts encouraged pastoralists, hitherto an ethnicized category of occupational specialists known as Bahima (a term used throughout the region), to regard themselves instead as “Tutsi,” an elite stratum of herders tied to royal power. At the same time, farmers were subjected to novel forms of surplus extraction for the benefit of Tutsi overlords. Thus, what had formerly been vertical categories of Bahima and farmers became transformed, gradually and unevenly, into hierarchical relations between “Tutsi” and “Hutu.” The latter ethnonym originated as a term of abuse, meaning uncivilized bumpkin, which Tutsi directed at their menials.Footnote ²⁶

These trends were intensified and transformed after 1897 by colonial rulers who governed via a Tutsi elite that they understood in terms of Western race science. But the era's most consequential historical narratives were written by Tutsi intellectuals who, in addition to their mission education, were adepts in the sophisticated oral historiography that had flourished at the courts of the nineteenth-century Rwandan kings. Those dynastic histories provided templates for narratives of Tutsi state-building that incorporated Hamitic motifs. Their grounding in the dynastic histories also supplied them with the kind of authority that ensured they would be taken seriously by critical Rwandan audiences that respected the precolonial intellectual traditions.Footnote ²⁷ The Tutsi intellectuals’ narratives were later taken up by Hutu activists, who recast them as tales of racial oppression.Footnote ²⁸

So, although Western teachings were of undoubted significance to this story of racial polarization, they were only one strand among many that Rwandan thinkers spun into historical narratives of racial difference, the other strands being inherited, not borrowed. While the language of race-science was indeed novel, other concepts learned in Belgian classrooms had Rwandan parallels. These included the language of difference via descent: Rwandans described the categories of Hutu and Tutsi as ubwoko, an ancient noun that designates shared common descent (it is often translated as “race” or “clan”).Footnote ²⁹ Rwandans were also familiar with a precolonial language of civilization and barbarism, which was often couched in terms of inheritance. That is hardly surprising: throughout world history, civilizational discourses have been central to the self-image of expansionary states like nineteenth-century Rwanda. In China, Southeast Asia, and elsewhere, conquest brought civilization, and the barbarians who remained outside its domain or who were forced unwillingly into it represented a different, unimproved category of humanity.Footnote ³⁰ Such indigenous ideas resonated with those of Western stadial historicism.

The political scientist Mahmood Mamdani argues that by casting the Tutsi as an alien ruling race, Belgian rule was a prime example of how Western colonizers constructed African political identities of indigenous and non-indigenous.Footnote ³¹ But, as I've indicated, there was more to it than that. Rwanda's dynastic historians built on rich discursive traditions that characterized political authority in terms of indigeneity and exogeny. Those traditions were older than the Rwandan kingdom itself. Throughout central and southern Africa, including in places where colonial rulers never espoused theories of “Hamitic” ruling races, one finds royal dynasties whose founding myths speak of civilizing outsiders. These exogenous civilizers are often contrasted with mythical figures of autochthonous nature: that is, inhabitants understood to have sprung from the local soil. The autochthons’ descendants are sometimes thought to still be around.

In short, the modernist focus on colonial officials and educators is not so much incorrect as it is incomplete. It flattens intellectual history into one where Africans have little agency save to adopt colonial ideas. I prefer instead to approach racial thought in the same manner as scholars approach other topics in intellectual history, where Africans’ encounter with Western discourses took the form not of an embrace but an entanglement.Footnote ³² Of course, Tutsi and Hutu racial thinkers had been influenced by Western ideas. But Tutsi supremacy was not a colonial invention. In Africans’ intellectual engagement with the West, borrowed ideas became entangled with discourses of difference they had inherited from their precolonial past.

Any history of racial thought within African intellectual traditions must take account of those inherited discourses. What form did they take? How, precisely, did modern racial thinkers make use of them? And, did they ever take racial forms in the centuries before European conquest? In the following pages, I will describe a few examples of locally inherited discourses that have, at times, shown signs of becoming racialized. Such discourses took a variety of forms, including those that a casual observer might describe as “class,” “caste,” or “ethnicity.” But I will focus on those that suggest the presence of African historicisms that arranged “human kinds” along a progression from barbarian to civilized.

*****

The most extensively studied African examples of civilizational discourses derived their core ideas from the Abrahamic faiths. This includes Ethiopia, where rulers of the expansionist Christian kingdoms in the Semitic-speaking imperial core developed perhaps the closest thing to an indigenous version of the Hamitic myth. The Abyssinian rulers were said to be descended from “the seed of Shem,” via the first-born son of King Solomon. These myths were enshrined in the Kebra Negast, the early fourteenth-century compilation that emphasized not only the ruling line's Solomonic ancestry (including genealogical ties to Christ) but also its destiny to rule over less exalted peoples on the peripheries of the Semitic-speaking highlands.Footnote ³³ Most degraded of these “savages who did not acknowledge God” were the so-called shankilla, Nilo-Saharan-speaking pastoralists who lived on and beyond the empire's lowland marches, where plow agriculture was not feasible. Unlike other peripheral people, shankilla could only be incorporated into the Christian polity as slaves. For centuries highlanders had imagined their degradation could be read in the “blackness” of their bodies.Footnote ³⁴

Of wider geographical distribution were civilizational motifs adapted from Near Eastern Islam.Footnote ³⁵ Written sources produced in West Africa document the presence of ideas that equated barbarism and unbelief with ancestral origins in the Bilad al-Sudan, the land of the blacks—ideas that, conversely, prompted Sudanic state-builders and other elites to imagine Arab ancestries for themselves. Under colonialism, those discourses meshed with European racial ideas, including historicist ideas that attributed governing skills to the more “civilized” among colonial subjects. In instances where colonial administrators ruled through what they called “native states,” such as the emirates of Hausa-speaking northern Nigeria, the expediency of indirect rule was complemented by civilizational discourses that seemed confirmed by local historical narratives. The emirates’ rulers were Fulbe by ancestry, historically pastoralists who had ranged throughout the grasslands of the western Sudan. Early in the nineteenth century, Fulbe religious scholars led a jihadist movement that conquered the Hausa city-states, knitting them together in a loose caliphate headquartered at Sokoto. Although by the century's end the Fulbe aristocrats spoke the same language as their subjects and practiced the same culture, they justified their social and political superiority by a belief that they had deeper genealogical roots in Islam than did their Hausa subjects; indeed, they believed that their paternal ancestors were Arabs, Companions of the Prophet.Footnote ³⁶ These civilizational discourses, tightly intertwined with discourses of descent, appealed to the parallel notions of Nigeria's new British overlords. In a fascinating wrinkle described by Moses Ochonu, the British even thought that the emirs’ Hausa subjects, although lacking the Fulbe's pedigree, had so benefited from their long exposure to Fulbe civilization that they could make fit rulers over the more barbaric peoples of what had been the caliphate's southern marches. Hausa-Fulbe civilizational discourses of racial difference were thus used to justify violating one of the cardinal principles of the British philosophy of indirect rule, in a core region of its birth: the principle that African intermediaries should be selected from among each constituency's own native “tribe.”Footnote ³⁷

In regions with long experience of Islam, the prestige of Arab ancestry often meshed with the concept of jahiliya, the era of “ignorance” or “barbarism” that preceded the revelation of the Quran. This fostered a particular kind of civilizational discourse that was disdainful of cultural practices and people who were deemed of purely indigenous origin.Footnote ³⁸ But those concepts were not imported into an intellectual vacuum any more than were ideas later imported from the West. On the Swahili coast of East Africa, they became entangled with the ancient motifs, already mentioned, that attributed civilizing processes to exogenous intruders. By the sixteenth century, if not earlier, those motifs had taken a form in which the civilizers were remembered as having originated in distant places across the Indian Ocean—that is, in the Islamic heartland. For reasons not altogether clear, the founder-heroes were often said to have come from the Persian town of Shiraz. Vague claims of “Shirazi” or “Arab” descent, unsupported by precise genealogies, were guarded by members of the urban elite, who deployed Arab- and Islam-centered concepts of “civilization” and “barbarism” to repel outsiders’ claims to full membership in community institutions. In the eighteenth century these tensions were made more complex by the arrival at the coast of powerful Omani merchants and freebooters, who became important political patrons of the local elite, thus intensifying the prestige and currency of Arab cultural influences. These processes culminated in the nineteenth century with the establishment of a powerful sultanate at Zanzibar ruled by Omani princes and dominated by creolized families who married locally and presided over a plantation economy worked by slaves imported from the continental interior.Footnote ³⁹

Likewise in northern Nigeria, Fulbe conquerors’ narratives of their Arab ancestry paralleled pre-existing origin myths by which the Hausa nobility they had overthrown, the sarauta, traced their own origins to a foreign hero. This hero, after spending time in the ancient kingdom of Bornu (and, in some variants, in North Africa or Baghdad), arrived in Hausaland, where he married a daughter of the autochthonous ruling family. Their seven children later overthrew the autochthonous rulers and founded the seven classic city-states of Hausa tradition. Guy Nicolas writes that the Hausa aristocrats thus traced their noble essence to the “fact of [their] foreignness,” that is, to their descent from early conquering immigrants. It was this exogenous essence that distinguished the sarauta from the autochthonous population of the villages, a “secular” distinction later elaborated by the kind of Islamic discourses that dominated intellectual life during the nineteenth-century rule of the jihadist states.Footnote ⁴⁰ Similar processes can be found in Darfur.Footnote ⁴¹ In all these places, discourses of civilization, barbarism, and racial difference introduced from the Islamic Near East lay like palimpsests on a substrate of earlier ideas.

*****

The legacy of slavery was central to many of the discourses described above. Unbelief had been the classic justification for enslavement in Islam, and state-builders in Muslim West Africa had long enslaved war-captives, both for export across the Sahara and for local use. But wars of conquest provided temptations to apply post-facto determinations of who might properly be regarded as Muslim. So, conquerors deemed entire geographic regions as “Dar al-Harb,” the land of warfare and unbelief, and their entire populations as “zunuj,” enslaveable barbarians (a concept akin to the Abyssinian shankilla). In the sultanate of Darfur, these ideas were grafted onto much more ancient traditions that disdained people who lived outside the rule of the state. A nineteenth-century song scorned such people as “slaves who yet go free”; as heathens and cannibals who wander about naked.Footnote ⁴² The convergence of such ideas yielded particularly vivid understandings of barbarism as an inherited trait that could be read not only in behavior but also in the body. Among Fulbe intellectuals connected to the Sokoto Caliphate, the characterization of any state or people as black or Sudanese connoted “the general charge of heathenism.”Footnote ⁴³ But such ingrained attitudes could also be found in regions that had never been under Caliphate rule. Fulbe in Burkina Faso commonly believed that all non-Fulbe, whether enslaved or not, had inherited traits that rendered them innately servile. They were “black, fat, coarse, naïve, irresponsible,” and “dominated by their needs and their emotions”—the direct opposite of the ideal Fulbe.Footnote ⁴⁴

Muslim scholars did not uniformly accept these ideas; they debated them for generations. The sharpest debates arose from the unlawful enslavement of people who were, in fact, freeborn Muslims, an act sometimes justified on the grounds that their Islam was insufficiently orthodox.Footnote ⁴⁵ Critics often cited a seventeenth-century text by the Timbuktu scholar Ahmad Baba. On the central point, Baba's judgment was clear: Islam forbade the enslavement of Muslims, and skin-color made no difference in that regard. The detailed attention he gives to refuting queries that impute an inherent barbarism to black skin (many of his interlocutors mentioned the curse of Ham) attests to the wide circulation of those concepts among Berber and Arab slave traders. But when it comes to the enslavement of entire categories of peoples below this level of black/white, Baba's judgment is ambiguous. He is less concerned with the question of whether it is possible to characterize entire peoples as non-believers as much as he disagrees about which peoples can be so judged: while it was unlawful to enslave captives taken from peoples or polities whose rulers were known for having voluntarily embraced Islam, descendants of other “clans” or “nations” were fair game. Baba provided a “confessional ethnography” of such collective distinctions which later jurists found invaluable.Footnote ⁴⁶ So even for Baba, there was slippage between the categories of civilized belief and descent. But most pertinent was the middle ground accepted by all participants in these debates. Even the sternest critics, Baba included, accepted that conversion did not erase slave status; on the contrary, no matter how pious a slave, her servile status stemmed from her “original non-belief,” that is, her ancestors’ refusal to convert. The practice of slavery, in other words, reproduced and reinforced discourses that enshrined barbarism and unbelief as elements of descent.Footnote ⁴⁷

Although Islam provided a powerful cultural and legal discourse for limning distinctions between the civilized and the barbarian, it was not responsible for introducing slavery to Africa.Footnote ⁴⁸ And whether Muslim or not, any society with widespread and long-lasting institutions of slavery (what Moses Finley distinguished as “slave-societies”) might produce discourses of inherited difference that continued to “nourish a kind of everyday racism” vis-à-vis slave descendants long after emancipation.Footnote ⁴⁹ Highland Ethiopians boasted a terminology that stigmatized precise degrees of slave descent as far back as the seventh generation (that is, a person of one-128th part slave ancestry).Footnote ⁵⁰ In twentieth-century Zanzibar, stereotypes derived from the history of slavery were even imposed on voluntary labor migrants who had come long after abolition, which, in turn, prompted the migrants to identify with a history of oppression at the hands of “Arab” slave-traders.Footnote ⁵¹ By contrast, where slavery was only incidental (“societies with slaves”), it left only vague memories, usually in the form of unequal ties of kinship between “junior” slave and “senior” master lineages.Footnote ⁵² But even in such places, slave descendants might continue to be stigmatized because of their ancestry. In postcolonial Ghana, suspicion of slave ancestry can prevent appointment to an Akan chieftaincy.Footnote ⁵³

*****

One need not look only at centralized states or slave-societies to find discourses of inherited civilization and difference. They can be found in settings throughout the continent, as can signs of their incipient racialization and their entanglement with more explicit racial discourses introduced from the West or the Arab Middle East. Most common are historical motifs that distinguish between immigrants and autochthons, a discursive tradition mentioned above. Those motifs in turn are part of a broader and probably quite ancient political culture that conceives authority in terms of frontiers and pioneers—a political culture so widespread as to have prompted the anthropologist Igor Kopytoff to describe it as part of a continent-wide “cultural ecumene.” In their most general form, these narratives tell of pioneers on a frontier who built communities by welcoming later arrivals; the latecomers were incorporated as junior kin or through other arrangements in which they acknowledged the firstcomers’ authority and their prior claims to land and other resources. The possible permutations of such narratives were endless; what they held in common was an idiom in which history was told as a “code of arrivals.” In many cases the first pioneers are remembered as having migrated from a long-settled region or well-established polity that was emblematic of a “mature” civilization (as Kopytoff glossed these concepts).Footnote ⁵⁴ We have already seen traces of such motifs in the myths of the Hausa founding heroes who had come from Bornu or (under an Islamic, Arab-centered overlay) from Baghdad.Footnote ⁵⁵

The founding pioneersFootnote ⁵⁶ are not always remembered as having arrived on an empty landscape, and, depending on their relative numbers and power, they would have had to establish relations with people already present. The ambivalent nature of those relations is reflected in historical narratives, and in associated social and ritual practices, where the earliest residents are remembered as autochthonous, in the precise sense of that term: that is, as people whose origins tied them to the earth in a primal way. In many narratives, the autochthons are remembered for having been driven off (a common motif tells of them disappearing into the earth from which they had come), in which case the arriving pioneers inherited or appropriated some of their ritual powers. But in other cases, the present-day community still includes families acknowledged as being of autochthonous descent.

In societies where such concepts hold sway, the status of chronological primacy attributed to the autochthonous stratum is usually expressed in terms of their special ritual ties to the land. In the Mossi kingdom of Yatenga in Burkina Faso, for example, the population was divided between the moose, regarded as descendants of the exogenous founding-hero, and the tengabiise, descendants of the autochthons. The kingdom's exogenous founder-hero had introduced the arts of statecraft, and his descendants were imbued with the quality of creative political power, or naam: the ability to wield political force over others.Footnote ⁵⁷ The moose, then, were understood as the gens du pouvoir, in Michel Izard's translation of these Mossi concepts. But they took care to maintain structured relationships with the autochthonous families, the gens de la terre (or maîtres de la terre), whose inherited ties to nature and to the soil invested them with the ability to control fertility and the rains.Footnote ⁵⁸ Traces of tensions between similar categories can be found in the early histories of many polities, as exogenous founder-heroes were pressed by circumstances to co-opt or otherwise gain control over the autochthons’ ritual powers.Footnote ⁵⁹ The narratives sometimes tell this tale through a motif of matrilateral conflict: the autochthons welcome the founding pioneer and give him one of their daughters to wed; the children of that union, once grown, rise up and overthrow the rule of their autochthonous maternal grandparents, displacing them in the name of their paternal, exogenous line. We have seen another example in origin-myths of the Hausa sarauta.Footnote ⁶⁰

What we have here, then, are ranked ethnic categories whose members understand themselves as standing in a relationship stemming from a history of displacement or conquest. Understandably, modern authors go out of their way to refute colonial-era interpretations that depicted such categories as vestiges of ancient racial conquests, along the lines of the Hamitic hypothesis. But the categories themselves were significant in precolonial thought, and while we might hesitate to describe them with the word “race,” we should recognize the family resemblance. As such, these discourses were susceptible to becoming racialized, especially when they became entangled with others. The motif of matrilateral conflict between autochthons and immigrants, for example, appears in the Arab-centered foundation myths concerning the Fulbe emirates and the Swahili city-states.Footnote ⁶¹ In more recent times, discourses of autochthony and exogeny have become politicized and racialized through entanglement with the neoliberal land policies of the postcolonial nation-state. The anthropologist Jean-Pierre Dozon, among others, has shown how tensions arising from such policies in Ivory Coast got mapped onto local discourses of autochthons and pioneers—of gens de la terre and gens du pouvoir—that, in turn, can be traced to the late eighteenth century if not earlier.Footnote ⁶² Much of the ensuing bloodshed has been of the dramatic, theatrical kind characteristic of racial violence—violence, as I have argued elsewhere, that cannot be explained simply in instrumental terms, as is common among scholars who wish to minimize the role of racial or ethnic subjectivities.Footnote ⁶³

*****

Perhaps the most telling examples of quasi-historicist discourses of autochthony and difference pertain to distinctions between sedentary farmers and neighbors who specialize in hunting and foraging. In the 1980s and 1990s these matters gave rise to contentious debates in two separate literatures: on the Basarwa or “Bushmen” of southern Africa and the Batwa or “Pygmies” of the Congo basin rainforest.Footnote ⁶⁴ The conventional view, which was part of colonial knowledge, had regarded the foragers as racially distinct from the farmers: as living fossils who had failed to embark on the evolutionary stages of progress that led to settled agriculture. But revisionists noted, from both contemporary observations and historical sources, that foragers often depended on sustained relationships with farmers, developing specialized skills to provide niche products for exchange with farmers or undertaking part-time farm labor. In many ways, the debates that ensued from these two positions engaged old assumptions about ethnic categories as units of analysis; the revisionists who launched the so-called Kalahari Debate took their lead from a classic essay in which Shula Marks challenged the notion that San hunter-gatherers and Khoikhoi herders constituted discrete, ethnically circumscribed ways of life in the Dutch Cape Colony.Footnote ⁶⁵ To the extent that the revisionists have won this larger battle—that the ethnic “group” has been dismantled as a unit of historical or sociological analysis, and we recognize the African past to have been characterized by broad cultural ecumenes and inter-braided cultural practices—the questions in these debates have lost much of their urgency.Footnote ⁶⁶ But the revisionists advanced an extreme instrumentalist position that the “ethnic” distinctions were mere illusions, the result of conjunctural circumstances that had reduced former farmers to the level of landless “serfs,” forced to forage for want of better alternatives.

In fact, the evidence is overwhelming that, in many places, distinctions between farmers and foragers, and the symbiotic relations between them, have been maintained for generations, and have given rise to complex discourses expressive of their ambiguous relationship. In other words, they do not merely reflect ad hoc or conjunctural conditions as suggested by the instrumentalist view. In the rainforest of the Congo basin, farmers and “Pygmy” hunter-gatherers perceive themselves as belonging to two discrete ethnic communities and understand their differences in terms of descent and corresponding physical traits. Yet at the same time, they are closely tied to one another in tight-knit relationships based on economic specialization. The precise forms of those relationships have varied from place to place and have always been subject to historical change. But in many instances they have been structured and ongoing, reproduced over extended periods.Footnote ⁶⁷ Moreover, they have been ranked: although the relative valuation of each side's economic contribution depends on one's perspective, both generally agree that the foragers “depend” on the farmers for their main caloric intake. Despite this shared language of Pygmy dependency, the relationship is understandably ambiguous, marked by mutual disdain and, on the part of the Pygmies, the sense of frustrated hostility that Nietzsche called ressentiment, itself a sentiment that bespeaks a ranked relationship.Footnote ⁶⁸ Alongside their scorn, farmers fear and even respect Pygmies as masters of the forest environment who possess crucial skills and ritual powers that the farmers themselves lack.Footnote ⁶⁹ Those skills and powers are said to derive from Pygmies’ status as descendants of the forest's autochthonous inhabitants.

Evidence of longstanding ethnic distinctions between farmers and foragers is derived from many sources, including archaeology and historical linguistics. But the most emphatic arguments cite genetic evidence: as one paleoanthropologist notes, African hunter-gatherers have become “iconic in the genetic literature.”Footnote ⁷⁰ In a way, such scholarship seeks to determine the extent to which hunter-gatherers constitute a separate race, in the biological sense—and an “autochthonous” race, to boot, insofar as it also usually seeks to demonstrate the presence of this distinct foraging population millennia before farmers or any others arrived on the scene. Understandably, this can cause discomfort among historians, since it seems to reify scientific concepts of race that we have learned to distrust. But we should keep in mind that the scientific concepts are not altogether exceptional. Rather, they share with all concepts of racial and ethnic difference the core idea of sexual avoidance, that is, the assumption that sexual reproduction across boundaries is limited or altogether absent. Of course, the notion that racial boundaries can be determined biologically is subverted by the simple fact that human sexual behavior is always more “disorderly” than idealized categorical orders might imply.Footnote ⁷¹ Debates about population genetics in central and southern Africa, then, only divert attention from the main issue: no matter what the genetic evidence suggests, pervasive disapproval of intermarriage, or ideals that limit it to forager female hypergamy (in which forager women can marry “up”), reflect longstanding perceptions of genealogical separation among farmers and foragers themselves.Footnote ⁷²

These perceptions of genealogical separation were not introduced by colonialism; the evidence from historical linguistics indicates that they have existed for a long time.Footnote ⁷³ Most striking is how they conceive of hunter-foragers as autochthons whose way of life is a primordial element held over from ancient times, before farmers ever arrived on the scene. Those concepts are reflected in words that exist in many of the region's Bantu languages, cognates of “Batwa” and “Basarwa,” which signify autochthons who live in the forest by hunting and foraging or who originally did so. These words are ancient, as are their attested meanings.Footnote ⁷⁴ In her deep history of the west-central rainforest, Kairn Klieman contends that the “ideology of the primordial Batwa” has remained a “root metaphor” for centuries, “adjusted to fit major transformations in social, political, and intellectual thought.”Footnote ⁷⁵

Such indigenous concepts of primordial Pygmy autochthony parallel evolutionist historicism. The anthropologist Roy Grinker, who conducted research in the Ituri forest of northeastern Congo, has described in detail some motifs that are common among farmers’ perceptions of foragers. In a variation of a motif about autochthons that can be found throughout the continent, the foragers in this instance, known as Efe, are said to have originally emerged from the stumps of trees that were being cleared by newly arrived farmers, the Lese.Footnote ⁷⁶ At the time of Grinker's research in the 1980s, Lese and Efe lived in ongoing, tightly-knit symbiotic relations; each Lese family had a sustained tie with “its” Efe, whom it “fed” from its farm in exchange for bushmeat and other forest products. Yet the farmers considered Efe quintessential outsiders: while Lese were “village people,” Efe were “forest people.” Lese used the metaphor of marriage to describe the relationship: like a wife in her husband's village, Efe are permanent outsiders. And as the farming village was the locus of civilization, farmers regarded Efe as emblematic barbarians. In the Lese moral universe, as among farmer communities throughout the broader region, civil order inheres in discipline and control, including the ability to clear forest for farming: qualities that John Lonsdale, in another context, transcribes as self-mastery.Footnote ⁷⁷ Those are qualities which, in the farmers’ view, the foragers lack. “The Lese view the Efe as they view the forest,” writes Grinker: “both are wild.” Forest people are unable to harness their passions or control their defecation; they copulate with abandon.Footnote ⁷⁸

But although they lacked the farmers’ qualities of sibosibo (foresight, patience, and rationality), the autochthonous Efe were respected by the farmers for their knowledge of matters pertaining to nature. Like autochthons elsewhere (including gens de la terre in the relationships described by Izard and others), Efe provided their farming patrons with ritual services connected to the territorial spirits that ensured fertility and fecundity. The nature of these complementary knowledges is captured in a widely told myth about the origin of the Lese-Efe relationship. An Efe ancestor was bewildered by the behavior of his grandmother, who, after defecating, habitually wiped her anus on his thigh. His Lese trading-partner showed him how to stop such savagery, through an act of shrewdly calculated, socially useful violence. The Efe, in gratitude, taught the farmer how to copulate and reproduce. Efe autochthons thus taught the farmers about the life-giving forces that stemmed from animal urges, nature, and the forest; the Lese newcomers taught the forest people how to control those forces.Footnote ⁷⁹

Although these images are specific to central Africa, the ideas they reflect should strike a chord in anyone familiar with Western racial thought. Their central component, as of others that contrast autochthons and civilizing newcomers, might be described as an African version of evolutionist historicism, in which Pygmies and other autochthons figure as living remnants of earlier, pre-civilized humankind. Such concepts, of course, resemble how Hegel or Trevor-Roper characterized all Africans. But a more illuminating comparison is with an evolutionist writer who actually visited Africa, Joseph Conrad. As Marlow travels up the river in Heart of Darkness, toward the very region Grinker describes, he felt he was going back to the beginnings of the world. Still, he could not restrain his admiration for the qualities of the Africans he encountered: their wildness, their closeness to nature. (Such evolutionist thought continues to echo in stereotyped praise of Africans’ presumably natural abilities in dance and other intuitive creative endeavors.) Marlow mocked the Europeans in the Congo, himself included, as mere phantoms, passing through; Africans, in contrast, belonged where they were, as if rooted to the soil. Pygmies in another part of the rainforest, supposed autochthons, likewise regard “village people” as only “passing by.” Farmers “came into life,” they like to say, “but will go again.”Footnote ⁸⁰

Is this African historicism merely a reflection of Western influence? To be sure, there is ample evidence that contemporary Africans apply Western-derived concepts to denigrate hunter-foragers as unlettered, unlearned, and resistant to progress.Footnote ⁸¹ But in their specifics, the African stereotypes differ sufficiently from the European ones to make one doubt a simple, unilinear story of diffusion. The common European fixation on Pygmies’ short stature, for example, does not figure in the African stereotypes Grinker describes. The evidence from historical linguistics suggests that the motifs described by Grinker and others can be traced to times long before Western intellectual impact. So Klieman's evaluation is no doubt correct: there were “two primordialist paradigms” about Batwa difference, one Western and one African. In the history of autochthony and historicism, as in the history of barbarism and civilization, Western and African intellectual interaction was a story of convergence and entanglement, not indoctrination.

*****

In suggesting the possibility of African historicist discourses and their convergence with Western notions of racial difference, I follow authors who argue that we take seriously African thinkers’ role in the creation of European “Africanist” knowledge.Footnote ⁸² To be sure, modern ethnic thought in Africa shows the influence of Western ideas, to say nothing of the social and political contexts created by colonial rule and colonial and postcolonial capitalism: this central contention of the modernist narratives cannot be denied. But in tracing the emergence of modern ethnic thought, one should not ignore the legacy of older, local ways of thinking about difference. The literature on precolonial Africa, including longue durée studies of early history, reveals the existence of indigenous discourses of difference that utilized metaphors of descent. Modern histories show how those discourses were often transformed under twentieth-century circumstances into “tribe” and “race.” But the histories of those discourses started long before the colonial moment. Certainly, there are no grounds for posing categorical distinctions between modes of ethnic thought and insisting that one mode (e.g., race) has purely European origins. Labels such as “race,” “ethnicity,” and “nationalism” are useful only as descriptive devices, not analytical categories, and insisting on a firm boundary between them can only impede historical investigations into how a form of thought we are content to describe with one of these labels may have emerged from something that looked quite different. The central component shared by all these modes of thought—the propensity to categorize human kinds via metaphors of common descent—seems to be ubiquitous in world history, not restricted to any one part of the globe.Footnote ⁸³

A telling illustration of the protean nature of ethno-racial thought can be seen by observers of current African affairs, as the boundaries of some of Africa's postcolonial nation-states seem to be in the process of becoming racialized. In theory, this should not have happened: African nationalisms were resolutely civic (or at any rate based on pan-Africanist notions of belonging that were so broad as to have allowed for the easy sway of jus soli), and the states’ boundaries had been drawn by European consuls who were famously oblivious to pre-existing political or ethnic identities.Footnote ⁸⁴ Thus many observers were confident that African nationalisms would prove different from those that had produced so much brutality in modern Europe.Footnote ⁸⁵ But in many parts of the continent, the past twenty-five years have seen the rise of “new nativisms” that take the nation-state itself as an “object of devotion” and have prompted popular violence against immigrants from other African countries.Footnote ⁸⁶ Perhaps the best-known instances have been in Ivory Coast and South Africa. In the latter, the targets of popular violence have included people who have been continuing the more than century-old patterns of labor migration on which the wealth of modern South Africa was built. They have been vilified as barbarians or makwerekwere (the word supposedly indicates the gibberish sounds they make instead of proper language, thus echoing the classic etymology of the Greek barbaros), and as carriers of witchcraft, crime, and contagion, in particular HIV. Given that the recently overthrown system of apartheid had been based on persistent efforts to engineer divisions of nation, race, and tribe, the ironies of this situation have troubled the moral imaginations of many South Africans.Footnote ⁸⁷

Given the recent vintage of Africa's national borders, most studies of these phenomena understandably emphasize their modern context, explaining them in terms of late twentieth-century globalization and the attendant intensification of labor flows and other forms of economic migration.Footnote ⁸⁸ Yet, valuable as they are, such explanations are not in themselves sufficient. In Ivory Coast, rhetoric directed against immigrants from Mali and Burkina Faso has been shaped by a much deeper history of tension between Muslim merchants from the north and people of the forest and forest-savanna transition zones in the south. That fellow citizens from northern Ivory Coast (and, at times, all Ivorian Muslims) are falsely denigrated as Burkinabè foreigners points to how the new nativism has become imbricated with those older discourses. Another source of nativist conflict in Ivory Coast involves tensions over neoliberal politics of “development” (mise en valeur). As I have indicated above, these tensions, too, have become racialized in part through their entanglement with much older inherited discourses of autochthony and civilizing pioneers. The racialized violence that has arisen from both axes of tension—in a country that, like post-apartheid South Africa, was once considered a textbook example of African civic nationalism—suggests that African nationalisms contain no fewer perils in this regard than their counterparts elsewhere in the world.