Despite all I'd read about Africa, my first impressions upon being there were overwhelming. As I walked the streets of Windhoek, the capital of newly independent Namibia, I saw black Herero people and black Ovambo; I saw Nama, a group quite unlike the blacks in appearance; I saw whites, descendants of recent European immigrants; and outside Windhoek I saw the last of the formerly widespread Kalahari Bushmen struggling for survival. These people were no longer pictures in a textbook; they were living humans, right in front of me. But what most surprised me was a street sign on one of downtown Windhoek's main roads. It read GOERING STREET.
Surely, I thought, no country could be so dominated by unrepentant Nazis that it would name a street after Hermann Goering, the notorious head of the Luftwaffe. As it turned out, the street actually commemorates Hermann's father, Heinrich, founding Reichskommissar of the German colony of South-West Africa, which would later be renamed Namibia. But Heinrich is no less a problematic figure than his son: his legacy includes one of the most vicious attacks ever carried out by European colonists on Africans, Germany's 1904 War of Extermination against the Herero. Today, while events in neighboring South Africa command the world's attention, Namibia, too, struggles to deal with its colonial history and establish a multiracial society. Namibia illustrated for me how inseparable Africa's past is from its present.
Most Americans think of native Africans as black and of white Africans as recent intruders; and when they think of Africa's racial history they think of European colonialism and slave trading. But very different types of peoples occupied much of Africa until as recently as a few thousand years ago. Even before the arrival of white colonialists, the continent harbored five of what many consider to be the world's six major divisions of humanity, the so-called human races, three of which are native to Africa. To this day nearly 30 percent of the world's languages are spoken only in Africa. No other continent even approaches this human diversity, and no other continent can rival Africa in the complexity of its human past.
The diversity of Africa's peoples results from its diverse geography and long prehistory. Africa is the only continent to extend from the northern to the southern temperate zone; it encompasses some of the world's driest deserts, largest tropical rain forests, and highest equatorial mountains. Humans have lived in Africa far longer than anywhere else: our remote ancestors originated there some 7 million years ago. With so much time, Africa's peoples have woven a complex, fascinating story of human interaction, a story that includes two of the most dramatic population movements of the past 5,000 years: the Bantu expansion and the Indonesian colonization of Madagascar. All those interactions are now tangled up in politics because the details of who arrived where before whom are shaping Africa today.
How did the five divisions of humanity in Africa get to be where they are today? Why did blacks come to be so widespread, instead of one or more of the four other groups whose existence Americans tend to forget? How can we ever hope to wrest the answers to these questions from Africa's past without written evidence of the sort that taught us about the spread of the Roman Empire?
African prehistory is a detective story on a grand scale, still only partly solved. Clues can be derived from the present: from the peoples living today in Africa, the languages they speak, and their plant crops and domestic animals. Clues can also be dug up from the past, from the bones and artifacts of long-dead peoples. By examining these clues one at a time and then combining all of them, we can begin to reconstruct who moved where at what time in Africa, and what let them move — with enormous consequences for the modern continent.
As I mentioned, the Africa encountered by the first European explorers in the fifteenth century was already home to five human races: blacks, whites, Pygmies, Khoisan, and Asians. The only race not found in Africa is the aboriginal Australians and their relatives.
Now, I know that classifying people into arbitrary races is stereotyping. Each of these groups is actually very diverse, and lumping people as different as the Zulu, Masai, and Ibo under the single heading "blacks" ignores the differences between them. So does lumping Africa's Egyptians and Berbers with each other and with Europe's Swedes under the single heading "whites." The divisions between blacks, whites, and the other major groups are arbitrary anyway because each group shades into the others. All the human groups on Earth have mated with humans of every other group they've encountered. Nevertheless, recognizing these major groups and calling them by these inexact names is a shorthand that makes it easier to understand history. By analogy, it's also useful to divide classical music into periods like "baroque," "classical," and "romantic," even though each period is diverse and shades into other periods.
By the time European colonialists arrived, most of Africa's major population movements had already taken place (see map on next page). Blacks occupied the largest area, from the southern Sahara to most of sub-Saharan Africa. The ancestors of most African Americans came from Africa's western coastal zone, but similar peoples occupied East Africa as well, north to the Sudan and south to the southeast coast of South Africa. They were mostly farmers or herders, as were the native African whites, who occupied Africa's northern coastal zone and the northern Sahara. (Few of those northern Africans — the Egyptians, Libyans, and Moroccans, for instance — would be confused with a blond, blue-eyed Swede, but they're often considered white because they have lighter skin and straighter hair than the peoples to the south.)
At the same time, the Pygmies were already living in groups widely scattered through the central African rain forest. Although they were traditionally hunter-gatherers, they also traded with or worked for neighboring black farmers. Like their neighbors, the Pygmies are dark- skinned and have tightly curled hair, but that hair is more thickly distributed over their body and face. They also are much smaller in size and have more prominent foreheads, eyes, and teeth.
The Khoisan (pronounced COY-san) are perhaps the group least familiar to Americans today. In the 1400s they were actually two groups, found over much of southern Africa: large-statured Khoi herders, pejoratively known as Hottentots, and smaller San hunter-gatherers, pejoratively called Bushmen. Most of the Khoi populations no longer exist; European colonists shot, displaced, or infected many of them, and the survivors interbred with Europeans. Though the San hunter-gatherers were similarly shot, displaced, and infected, a dwindling number managed to preserve their distinctness in Namibian desert areas unsuitable for agriculture. (They're the people depicted some years ago in the widely seen film The Gods Must Be Crazy.) The Khoisan today look quite unlike African blacks: they have light brown skin sometimes described as yellow, and their hair is even more tightly coiled.
Of these population distributions, that of North Africa's whites is the least surprising because physically similar peoples live in adjacent areas of the Middle East and Europe. Throughout recorded history people have been moving back and forth between Europe, the Middle East, and North Africa. But the puzzling placements of blacks, Pygmies, and Khoisan hint at past population upheavals. Today there are just 200,000 Pygmies scattered amid 120 million blacks. This fragmentation suggests that Pygmy hunters lived throughout the equatorial forests until they were displaced and isolated into small groups by the arrival of black farmers. Similarly, the Khoisan area of southern Africa is surprisingly small for a people so distinct in anatomy and language. Could the Khoisan as well have been originally more widespread until their more northerly populations were somehow eliminated?
Perhaps the greatest puzzle, however, involves the island of Madagascar, which lies just 250 miles off the coast of southeastern Africa, much closer to Africa than to any other continent. It's in Madagascar that the fifth African race is found. Madagascar's people prove to be a mixture of two elements: African blacks and — surprisingly, given the separation seemingly dictated by the whole expanse of the Indian Ocean — Southeast Asians, specifically Indonesians. As it happens, the language of the Malagasy people is very close to the Ma'anyan language spoken on the Indonesian island of Borneo, over 4,000 miles away. No one even remotely resembling the Borneans lives within thousands of miles of Madagascar.
These Indonesians, their language, and their modified culture were already established on Madagascar by the time it was first visited by Europeans in 1500. To me this is the single most astonishing fact of human geography in the whole world. It's as if Columbus, on reaching Cuba, had found it occupied by blue-eyed, towheaded Scandinavians speaking a language close to Swedish, even though the nearby North American continent was inhabited by Indians speaking Indian languages. How on earth could prehistoric people of Borneo, presumably voyaging in boats without maps or compasses, have ended up in Madagascar?
The case of Madagascar shows how peoples' languages, as well as their physical appearance, can yield important clues to their origins. Similarly, there's much to be learned from African languages that can't be gleaned from African faces. In 1963 the mind-boggling complexities of Africa's 1,500 languages were simplified by the great linguist Joseph Greenberg of Stanford. Greenberg recognized that all those languages can be divided into just four broad families. And, because languages of a given language family tend to be spoken by distinct peoples, in Africa there are some rough correspondences between the language families and the anatomically defined human groups (see map at right). For instance, Nilo-Saharan and Niger-Congo speakers are black, and Khoisan speakers are Khoisan. Afro-Asiatic languages, however, are spoken by a wide variety of both whites and blacks. The language of Madagascar belongs to yet another, non-African category, the Austronesian language family.
What about the Pygmies? They're the only one of Africa's five races that lacks a distinct language: each band of Pygmies speaks the language of its neighboring black farmers. If you compare a given language as spoken by Pygmies with the same language as spoken by blacks, however, the Pygmy version contains unique words and, sometimes, distinctive sounds. That makes sense, of course: originally the Pygmies, living in a place as distinctive as the equatorial African rain forest, must have been sufficiently isolated to develop their own language family. Today, however, those languages' disappearance and the Pygmies' highly fragmented distribution both suggest that the Pygmy homeland was engulfed by invading black farmers. The remaining small bands of Pygmies adopted the invaders' languages, with only traces of their original languages surviving in a few words and sounds.
The distribution of Khoisan languages testifies to an even more dramatic engulfing. Those languages are famously unique — they're the ones that use clicks as consonants. All the existing Khoisan languages are confined to southern Africa, with two exceptions: the click-laden Hadza and Sandawe languages spoken in Tanzania, some 1,500 miles from their nearest linguistic kin.
In addition, clicks have made it into a few of the Niger-Congo languages of southern Africa, such as Zulu and Xhosa (which is the language of Nelson Mandela). Clicks or Khoisan words also appear in two Afro-Asiatic languages spoken by blacks in Kenya, stranded even farther from the Khoisan peoples of today than are the Hadza and Sandawe speakers of Tanzania. All this suggests that Khoisan languages and peoples formerly extended far north into Africa until the Khoisan, like the Pygmies, were engulfed by the blacks, leaving behind only a linguistic legacy to testify to their former presence.
Perhaps the most important discovery from linguistic sleuthing, however, involves the Niger-Congo language family, which today is spread all over West Africa and most of subequatorial Africa. Its current enormous range seems to give no clue as to precisely where the family originated. However, Greenberg has pointed out that the Bantu languages of subequatorial Africa, once thought to be their own language family, are actually a subfamily of the Niger-Congo language family. (Technically they're a sub-sub-sub-sub-sub-sub-sub-sub-sub-subfamily.) These Bantu languages today account for nearly half of the 1,032 Niger-Congo languages, and Bantu speakers account for more than half (nearly 200 million) of the Niger-Congo speakers. Yet all 494 Bantu languages are so similar to one another that they've been facetiously described as 494 dialects of a single language.
There are some 170 other such Niger-Congo subfamilies, most of which are crammed into West Africa, a small fraction of the entire Niger- Congo range. Even the most distinctive Bantu languages, as well as the Niger-Congo languages most closely related to Bantu, are concentrated there, in a tiny area of Cameroon and adjacent east and central Nigeria.
From Greenberg's evidence it seems obvious that the Niger-Congo language family arose in West Africa, while the Bantu subfamily arose at the east end of that range, in Cameroon and Nigeria, and then spread out over most of subequatorial Africa. That spread must have begun sufficiently long ago that the ancestral Bantu language had time to split into 494 daughter languages, but nevertheless recently enough that all those daughter languages are still very similar to one another. Since all Niger- Congo speakers — including the Bantu speakers — are black, it would be nearly impossible to infer who migrated in which direction just from the evidence of physical anthropology.
To make this type of linguistic reasoning clear, let me give you an example: the geographic origins of the English language. Today the largest number of people whose first language is English live in North America, with others scattered over the globe in Britain, Australia, New Zealand, and other countries. If we knew nothing else about language distribution and history, we might have guessed that the English language arose in North America and was carried overseas by colonists.
But we know better: we know that each of those countries has its own English dialect and that all those English dialects make up just one subgroup of the Germanic language family. The other subgroups — the various Scandinavian, German, and Dutch languages — are crammed into northwestern Europe. Frisian, the Germanic language most closely related to English, is stuck in a tiny coastal area of Holland and western Germany. Hence a linguist would immediately deduce — correctly — that the English language arose on the northwestern coast of Europe and spread around the world from there.
Essentially the same reasoning tells us that the nearly 200 million Bantu-speaking people now flung over much of the map of Africa arose in Cameroon and Nigeria. Thus linguistics tells us not only that the Pygmies and the Khoisan, who formerly ranged widely over the continent, were engulfed by blacks; it also tells us that the blacks who did the engulfing were Bantu speakers. But what it can't tell us is what allowed the Bantu speakers to displace the Pygmies and Khoisan.
To answer that question we need to look at a different type of surviving evidence, that of domesticated plants and animals. Why is this evidence so crucial? Because farming and herding yield far more calories per acre than does hunting wild animals or gathering wild plants. As a result, population densities of farmers and herders are typically at least ten times those of hunter-gatherers. That's not to say that farmers are happier, healthier, or in any way superior to hunter-gatherers. They are, however, more numerous. And that alone is enough to allow them to kill or displace the hunter-gatherers.
In addition, human diseases such as smallpox and measles developed from diseases plaguing domestic animals. The farmers eventually become resistant to those diseases, but hunter-gatherers don't have the opportunity. So when hunter-gatherers first come into contact with farmers, they tend to die in droves from the farmers' diseases (see "The Arrow of Disease," October 1992).
Finally, only in a farming society — with its stored food surpluses and concentrated villages — do people have the chance to specialize, to become full-time metalworkers, soldiers, kings, and bureaucrats. Hence the farmers, and not the hunter-gatherers, are the ones who develop swords and guns, standing armies, and political organization. Add that to their sheer numbers and their germs, and it's easy to see how the farmers in Africa were able to push the hunter-gatherers aside.
But where in Africa did domesticated plants and animals first appear? What peoples, by accident of their geographic location, inherited those plants and animals and thereby the means to engulf their geographically less-endowed neighbors?
When Europeans reached sub-Saharan Africa in the 1400s, Africans were growing five sets of crops. The first set was grown only in North Africa, extending as far as the highlands of Ethiopia. North Africa's rain falls mostly in the winter months — the region enjoys a Mediterranean climate — so all its original crops are adapted to germinating and growing with winter rains. Archeological evidence tells us that such crops — wheat, barley, peas, beans, and grapes, to name a few — were first domesticated in the Middle East around 10,000 years ago. So it makes sense that they would have spread into climatically similar and adjacent areas of North Africa, laying the foundation for the rise of ancient Egyptian civilization. Indeed, these crops are familiar to us precisely because they also spread into climatically similar and adjacent areas of Europe — and from there to America and Australia — and became some of the staple crops of temperate-zone agriculture around the world.
There's little rain and little agriculture in the Sahara, but just south of the desert, in the Sahel zone, the rain returns. The Sahel rains, however, fall in the summer. So even if winter-rain-adapted Middle Eastern crops could somehow have crossed the Sahara, it would still have been hard to grow them in the summer-rain Sahel zone. Instead, here the Europeans found the second and third sets of African crops, both of which are adapted to summer rains and the area's less variable day length.
Set number two is made up of plants whose ancestors were widely distributed from west to east across the Sahel zone and were probably domesticated there as well. They include sorghum and pearl millet, which became the staple cereals of much of sub-Saharan Africa, as well as cotton, sesame, watermelon, and black-eyed peas. Sorghum proved so valuable that it is now grown in hot, dry areas on all the continents.
The wild ancestors of the third set of African crops are found only in Ethiopia and were probably domesticated there. Indeed, most of them are still grown only there: few Americans have ever tasted Ethiopia's finger millet beer, its oily noog, its narcotic chat, or its national bread, which is made from a tiny-seeded cereal called teff. But we all have the ancient Ethiopian farmers to thank for the domestication of a plant we know exceedingly well: the coffee plant, which remained confined to Ethiopia until it caught on in Arabia and then spread around the globe.
The fourth set of African crops was domesticated from wild ancestors in the wet climate of West Africa. Some of them, including African rice, have remained virtually confined there; others, such as African yams, eventually spread throughout much of sub-Saharan Africa; and two, the oil palm and the kola nut, spread to other continents. West Africans were chewing the caffeine-containing kola nut as a stimulant long before the Coca-Cola Company enticed Americans to drink its extracts.
The plants in the last batch of African crops are also adapted to wet climates. Bananas, Asian yams, and taro were widespread in sub-Saharan Africa when the Europeans arrived, and Asian rice was well established on the coast of East Africa. But these crops didn't come from Africa. They came from Southeast Asia, and their presence in Africa would be astonishing if the presence of Indonesians in Madagascar hadn't already alerted us to Africa's prehistoric Asian connection.
Let's consider the four indigenous groups of crops. All four — from North Africa, the Sahel, Ethiopia, and West Africa — came from north of the equator. No wonder the Niger-Congo speakers, people who also came from north of the equator, were able to displace Africa's equatorial Pygmies and subequatorial Khoisan peoples. The Khoisan and the Pygmies weren't unsuited for the farming life; it was just that southern Africa's wild plants were unsuitable for domestication. Even the Bantu and the white farmers, heirs to thousands of years of farming experience, have rarely been able to develop southern Africa's native plants into food crops.
Because there are so few of them, summarizing Africa's domesticated animal species is much easier than summarizing its plants. The list doesn't include even one of the big wild mammals for which Africa is famous — its zebras and wildebeests, its rhinos and hippos, its giraffes and Cape buffalo. The wild ancestors of domestic cattle, pigs, dogs, and house cats were native to North Africa but also to western Asia, so we can't be sure where they were first domesticated. The rest of Africa's domestic mammals must have been domesticated somewhere else because their wild ancestors occur only in Eurasia. Africa's sheep and goats were domesticated in western Asia, its chickens in Southeast Asia, its horses in southern Russia, and its camels probably in Arabia. The one exception is the donkey, which is widely believed to have been domesticated in North Africa.
Many of Africa's food staples and domesticated animals thus had to travel a long way from their point of origin, both inside and outside Africa. Some people were just luckier than others, inheriting suites of domesticable wild plant and animal species. We have to suspect that some of the "lucky" Africans parlayed their advantage into an engulfing of their neighbors.
But all the evidence I've presented thus far — evidence from modern human and language distributions and from modern crops and domestic animals — is only an indirect means to reconstruct the past. To get direct evidence about who was living where when, and what they were eating or growing, we need to turn to archeology and the things it turns up: the bones of people and their domestic animals, the remains of the pottery and the stone and iron tools they made, and the remains of the buildings they constructed.
This evidence can help explain at least some of the mystery of Madagascar. Archeologists exploring the island report that Indonesians arrived before A.D. 800, possibly as early as 300, and in a full-fledged expedition: the earliest human settlements on Madagascar include the remains of iron tools, livestock, and crops. This was no small canoeload of fishermen blown off course.
Clues to how this expedition came about can be found in an ancient book of sailors' directions, the Periplus of the Erythrean Sea, which was written by an anonymous merchant living in Egypt around A.D. 100. The merchant describes an already thriving sea trade connecting India and Egypt with the coast of East Africa. When Islam began to spread after the beginning of the ninth century, Indian Ocean trade became well documented archeologically by copious quantities of Middle Eastern and occasionally even Chinese products such as pottery, glass, and porcelain found in East African coastal settlements. The traders waited for favorable winds to let them cross the Indian Ocean directly between East Africa and India.
But there was an equally vigorous sea trade from India eastward, to Indonesia. Perhaps the Indonesian colonists of Madagascar reached India by that route, then fell in with the westward trade route to East Africa, where they joined with Africans and discovered Madagascar. The union of Indonesians and East Africans appears to live on today in Madagascar's basically Indonesian language, which contains loan words from coastal Kenyan Bantu languages. But there's a problem: there are no corresponding Indonesian loan words in Kenyan languages. Indeed, there are few Indonesian traces in East Africa besides some musical instruments like the xylophone and the zither and the Indonesian crops discussed earlier. Is it possible that the Indonesians, instead of taking the easier route to Madagascar via India and East Africa, somehow — incredibly — sailed straight across the Indian Ocean, discovered Madagascar, and only later got plugged into East African trade routes? We still don't know the answer.
The same sorts of archeological evidence found in Madagascar can be found on the African continent itself. In some cases they can help prove hypotheses that the other evidence could never fully resolve. For instance, linguistic and population distribution evidence merely suggests that the Khoisan were once widespread in the drier parts of subequatorial Africa. But archeologists in Zambia, to the north of the modern Khoisan range, have in fact found skulls of people resembling the modern Khoisan, as well as stone tools resembling those the Khoisan peoples were making in southern Africa when the Europeans arrived.
There are, of course, cases in which archeology can't help. We assume from indirect evidence that Pygmies were once widespread in the wet rain forest of central Africa, but it's difficult for archeologists to test this assumption: although they've found artifacts to show that people were there, they have yet to discover ancient human skeletons.
Archeology also helps us determine the actual dates and places for the rise of farming and herding in Africa, which, as I've said, is the key to understanding how one group of people was able to conquer the whole continent. Any reader steeped in the history of Western civilization would be forgiven for assuming that African food production began in ancient Egypt's Nile Valley, land of pharaohs and pyramids. After all, by 3000 B.C., Egypt was undoubtedly the site of Africa's most complex society. Yet the earliest evidence for food production in Africa comes not from the Nile Valley but from, believe it or not, the Sahara.
Archeologists are able to say this because they have become expert at identifying and dating plants from remains as fragmentary as charred seeds recognizable only under a microscope. Although today much of the Sahara is so dry that it can't even support grass, archeologists have found evidence that between 9000 and 4000 B.C. the Sahara was more humid; there were numerous lakes, and the desert teemed with game. The Saharans tended cattle and made pottery, then began to keep sheep and goats; they may even have started to domesticate sorghum and millet. This Saharan pastoralism began well before food production got its start in Egypt, in 5200 B.C., when a full package of western Asian winter crops and livestock arrived. Farming then spread to West Africa and Ethiopia. By around 2500 B.C. cattle herders had already crossed the modern border of Ethiopia into northern Kenya.
Linguistics offers another way to date the arrival of crops: by comparing words for crops in related modern languages that diverged from each other at various times in the past. It thus becomes clear, for instance, that the people who were domesticating sorghum and millet in the Sahara thousands of years ago spoke languages ancestral to modern Nilo- Saharan languages. Similarly, the people who first domesticated the wet- country crops of West Africa spoke languages ancestral to the modern Niger-Congo languages. The people who spoke ancestral Afro-Asiatic languages were certainly involved in the introduction of Middle Eastern crops into North Africa and may have been responsible for the domestication of crops native to Ethiopia.
Analyzing the names of crops leaves us with evidence that there were at least three ancestral languages spoken in Africa thousands of years ago: ancestral Nilo-Saharan, Niger-Congo, and Afro-Asiatic. And other linguistic evidence points to an ancestral Khoisan language (that evidence, however, doesn't come from crop names, since the ancestral Khoisan people didn't domesticate any crops). Surely, since Africa harbors 1,500 languages today, it was big enough to harbor more than four ancestral languages in the past. But all those other languages must have disappeared, either because the peoples speaking them lost their original languages, as the Pygmies did, or because the peoples themselves disappeared.