History of Foods: Africa South from the Sahara

James L Newman. Cambridge World History of Food. Editor: Kenneth F Kiple & Kriemhild Conee Ornelas. Volume 2. United Kingdom: Cambridge University Press, 2000.

Describing the principal sources of food for the inhabitants of Africa south from the Sahara is a relatively easy task. Most diets are dominated by products made from a single staple crop, and there are not all that many of them. Maize, sorghums, pearl or bulrush millet, and rice are the prominent grains, and cassava, yams, and bananas or plantains account for most of the vegetatively propagated varieties. Furthermore, their general geographies can be explained, for the most part, by annual totals and seasonality of rainfall. For example, near the dry margins of cropping, pearl millet makes its greatest dietary contribution, whereas the equatorial zone is where bananas and plantains come to the fore. Even adding in the role played by livestock, one that varies from insignificant to crucial, does not overly complicate the picture. Among farmers, fowl are fairly ubiquitous, while sheep, goats, and cattle are kept wherever diseases, especially sleeping sickness, do not prohibit them. When aridity intervenes to make crop cultivation too hazardous to rely upon, the herding of camels or cattle becomes the primary subsistence activity.

The problems come when attempting to go much beyond this level of generality. There is a plethora of other foods that are important to diets, including those from wild sources, and matters get even more difficult to sort out when issues of history, culture, and nutritional adequacy must be addressed. The region’s human diversity is enormous, and most food systems display a complex interweaving of influences, ranging from the distant past to the present. Unfortunately, trying to understand what has happened through time is hindered by a dearth of information. The written record is sparse before the twentieth century, and archaeology, so far, has produced very few dates. As a result, temporal insights often must rely on somewhat less precise sources of information, such as paleobotany, historic and comparative linguistics, and cultural anthropology.

Our understanding of current diets and their adequacy is only slightly better informed. During recent decades, Africa south from the Sahara has replaced China and India as the “land of famine.” Actual famine, however, has visited only restricted areas, such as portions of the Sahel, Somalia, and Mozambique. Elsewhere, people have not been starving, although only a few elites enjoy the luxuries of Western-like food abundance. Serious nutrition-related health problems clearly exist, especially among young children and reproductive-age women, and the food production data do point to an ever more precarious supply situation throughout much of the region. But surveys that allow for comparing diets and nutritional well-being are woefully inadequate.

As a result, the following presentation is offered in the spirit of being one best estimate. It begins with an overview of African “agricultural origins and dispersals,” to borrow a title from Carl Sauer (1969), which, contrary to many earlier interpretations, were not all by-products of Egyptian influences. Other Africans engaged in their own domesticating processes, leading to the crops that provided the basic raw materials for the transition to food-producing ways of life. The most significant imports, the subjects of the next section, entered already existing agricultural systems. A brief discussion of food preparations and eating habits follows, and the final section is devoted to an assessment of what appear to be the most widespread diet-related disorders, excepting famine.

African Agricultural Origins

We can begin some 9,000 years ago when all Africans southward from the Sahara still gathered, hunted, and fished for their foods, with the particular dietary mix governed by what the local environment could provide. For example, on the vast savanna-covered plateaus, the abundance of game animals allowed meat to be eaten regularly, along with a variety of nuts, berries, roots, and tubers. Meat was harder to come by in more densely forested habitats, but freshwater fish were widely available as complements to vegetable foods, whereas along some coastal sections, extensive middens testify to the dietary importance of shellfish. Virtually all of the area appears to have been occupied, at least on a seasonal basis; exceptions were the higher elevations, the heart of the Sahara Desert (much less extensive then than now), the narrow strip of the Namib Desert, and the deeper recesses of the equatorial rain forest. For the most part, people lived in small, mobile bands exploiting territories that necessitated the use of tens of square kilometers to supply the subsistence needs of one adult.

By way of contrast, certain riparian habitats north and east of the equatorial rain forest supported higher population densities and more sedentary occupancy. They did so because of Late Stone Age technological advances that allowed people to use a diverse array of plants, including wild grass seeds, for foodstuffs, as well as to engage in an intensified exploitation of fish and waterfowl (Sutton 1977). The evidence to date points to such locales as these as having been at the forefront of domestication activities in Africa south from the Sahara, thus lending support to the agricultural origins models developed by D. Rindos (1984) and R. S. MacNeish (1992). Shortly thereafter, other movements in the direction of agriculture took place along the rain forest-savanna ecotone of western and central Africa and within what is now Ethiopia.

Whether there were broad zones or precise centers of domestication activity at the outset cannot be determined, but no matter which situation prevailed, the diffusion pattern evolved into one of multiple frontiers carrying a variety of crops and animals. And the pace of change varied considerably. In some instances, Stone Age economies seem to have given way quickly, but the norm was for agriculture to gain ascendancy over gathering, hunting, and fishing only very gradually.


The earliest signs of possible domestication come from two sites within the Sahara. One is at Nabta Playa in western Egypt, where a number of sorghum-type seeds have been found in deposits dated 8,100 years ago (Wasylikowa et al. 1993). Whether they were cultivated or gathered, however, cannot be determined. The other is in the Ahaggar and contains a seed, two pollen grains, and grain impressions, dated between 6,850 and 5,350 years ago, of what could be pearl millet (Pennisetum glaucum) (Muzzolini 1993). This is not much to go on, but the Ahaggar does fall within the biogeographical range of related wild varieties, while the timing corresponds to a moist episode that would have supported millet cultivation at favorable Saharan locales, such as in the highlands and along the shores of then fairly numerous lakes and rivers (Harlan 1989). Also dated to approximately the same time as the Nabta Playa and Ahaggar finds are Neolithic sites that have yielded pottery, grindstones, and rubbers. Their presence by no means guarantees cultivation, but the large numbers of artifacts add extra fuel to the speculation that something of significance was indeed taking place (Camps 1982).

Farther south, events moved more slowly, although both botanical and linguistic research indicates that domestication processes were trending in the direction of agriculture within several different zones during the period from 5,000 to 4,000 years ago. One zone followed the Sahel-savanna country stretching from the upper Niger River valley, through the Lake Chad basin, to Dar Fur. Overall, the western half appears to have been more active, accounting for African rice (Oryza glaberrima), fonio (Digitaria exilis), guinea millet (Brachiaria deflexa), water-melon (Colocynthis citrullus),and Bambara ground-nuts (Voandzeia subterranea). The area east of Lake Chad was where the first sorghums (Sorghum bicolor) seem to have been cultivated. Later activities produced a variety of specialized sorghum races that would serve as the principle food source for peoples inhabiting the semiarid and subhumid regions of not only Africa but also India (Harlan 1989 and 1992).

So far, this vast zone to the south of the desert has yielded only two archaeological sites with substantiated evidence about the course of early crop cultivation. The oldest is Dhar Tichitt located in what is now south-central Mauritania, where P. J. Munson (1976) has documented a progression from an earlier riparian-type economy to a village-based one growing pearl millet by 3,100 years ago. No in-migration is indicated, thus pointing to diffusion from the Sahara as the likely source for the millet. The other site is Daima on the western side of Lake Chad. Here, excavations of earthen mounds have revealed the presence of a mixed economy involving fishing, hunting, gathering, herding, and crop cultivation centered on sorghums dating back to 2,500 years ago (Connah 1981). The agriculture practiced at Daima was advanced enough to support villages of more than 100 inhabitants and, therefore, must have been preceded by earlier developments that have not yet been unearthed.

Ever since the days of N. I.Vavilov (1926), the area within and around the Ethiopian highlands has been identified as a likely source of indigenous crop domestication. Now well established as coming from the cooler, seasonally rainy uplands of the center and east are teff (Eragrostis tef), noog (Guizotia abyssinica), and finger millet (Eleusine coracana), whereas ensete (Musa ensete) and coffee (Coffea arabica) are ascribed to the warmer, wetter southwestern forests. Dates are unconfirmed, but opinion based on linguistic analyses favors something on the order of 5,000 years ago for the upland crops (Ehret 1979; Phillipson 1993). This date may be fairly accurate because when wheat and barley arrived from south Arabian sources, about 3,000 years ago, they seem to have been incorporated into already well established agricultural systems.

Finally, various African communities appear to have been experimenting with yam cultivation (primarily Dioscorea rotundata and Dioscorea bulbifera ) along a wide front following the savanna-rain forest eco-tone in western Africa (Alexander and Coursey 1969). How far back in time this goes cannot be determined, but it must have been by at least 5,000 years ago because of the association of yams with several major population migrations dated to around this time (Ehret 1982). Because of the workings of humid climates and acidic soils on a plant that is inherently perishable to begin with, archaeology is unlikely ever to shed much light on the origins of yam cultivation, and, thus, further insights will have to continue to be based primarily on advances in botanical and linguistic research.

Kersting’s groundnuts (Kerstingiella geocarpa), cowpeas (Vigna unguiculata), and the oil palm (Elaeis guineensis) are other crops thought to derive from the same general ecotone environment (Harlan 1992), but so far only cowpeas have been archaeologically confirmed. These come from the several Kintampo sites focused on central Ghana, and are dated to around 3,800 years ago (Flight 1976). The inhabitants of Kintampo also made considerable use of oil palm kernels, but these probably came from the harvesting of stands of protected wild trees.


Unlike its crops, Africa south from the Sahara has depended upon external sources for its most important livestock. No possible ancestors for sheep or goats existed on the continent. Instead they arrived in several waves, from southwestern and central Asian sources, and quickly became integral parts of agricultural systems developed in Egypt and the Maghreb. The first cattle also seem to have entered from the east, although it is possible that indigenous varieties of Bos primigenius were domesticated within the Sahara (Smith 1992). These were all humpless types from which the fully domesticated Bos taurus arose. Later, but at least by 2,500 years ago, came humped Bos indicus types that probably entered Africa both via Egypt and from seaborne contacts with India along the east coast. Eventually, B. taurus and B. indicus crosses produced several distinctive African breeds, such as the Fulani, Sanga, and Afrikander (Blench 1993; Clutton-Brock 1993).

One initial route of livestock diffusion went upstream along the Nile, as is evidenced at the archaeological sites of Kadero and Esh Shaheinab near the junction of the White and Blue Niles. Wheat and barley may have been grown here as well, but it was sheep, goats, and cattle that constituted the main additions to a preexisting riverine-based Stone Age economy by 7,000 years ago (MacNeish 1992). From here, livestock continued to spread inland, reaching the region of Ethiopia between 5,000 and 6,000 years ago (Ehret 1979).

Roughly contemporaneous with Kadero and Esh Shaheinab are signs of other emergent livestock-based economies within and around the Saharan highlands of Dar Fur, Tibesti, Air, and Ahaggar. The evidence includes animal skeletons, material cultural remains, and, most convincingly, rock art at the remarkable site of Tassili-n-Ajer (Lhote 1959). Preserved there since approximately 7,000 years ago are scenes of what clearly are domesticated, humpless, long-horn cattle, along with their herders (Camps 1982).

When camels first made their way to Africa remains unclear. The dromedary had been domesticated in the Arabian peninsula by at least 3,800 years ago, but it does not seem to have become important in Egypt and the Sahara until much later, perhaps not until Roman times. Rather than entering with caravans crossing via the Sinai, as most experts have tended to think, camels may have first reached Africa by being transported across the Red Sea, a route which would help to explain their seemingly greater antiquity among desert-dwelling nomads in this corner of the continent (Zeuner 1963).

Even more mystery surrounds the entry of the common fowl (Gallus gallus) into Africa. We know that they were present in Egypt over 3,000 years ago, but that is about all (Simoons 1961). The usual assumption is that they probably spread from here up the Nile Valley corridor, and it is possible that chickens could have been carried across the Sahara once they had reached the Maghreb. A still later introduction (about A.D. 500) may have been with Malayo-Polynesian voyagers, who touched at points along the Indian Ocean coast on their way to Madagascar. Today, chickens are kept by farmers virtually everywhere, although much of their popularity seems to be quite recent.

Africa can claim two animal domesticates, neither of which, however, became important sources of food. One is the ass (Equus asinus), for which the Nubian wild race seems to have provided most of the genetic material (Zeuner 1963). It was in use as a beast of burden in Egypt, and presumably in Nubia, by 5,000 years ago, and from there spread to Ethiopia, the Horn, and the Sahel/Sudanic zone of western Africa at uncertain dates. The other is the guinea fowl, derived largely from Numidia meleagris galeata, which appears to have been first raised in the savannas of western Africa. Appreciated more for its aesthetic qualities than anything else, it often cannot be easily differentiated from the wild varieties that are still found widely outside the rain forest zones (Donkin 1991).

These are the essential domesticates that supported the rise of food-producing economies in Africa south from the Sahara. Over a span of 4,000 to 5,000 years, they have been carried across the subcontinent’s diverse environments by a complex mix of population migrations and technological diffusions. Many of these, however, can only be speculated about.

Regional Agricultural Dispersals

Western Africa

The pace of agricultural development in the Sahelian and Sudanic zones of western Africa seems to have intensified under competition and population pressure from southward-drifting livestock herders seeking respite from the ever-worsening aridity that had set in by 4,000 years ago. The evidence points to their having been members of a rapidly diverging Nilosaharan language group, other members of which were also the likely first cultivators of pearl millet (Ehret 1993). The most attractive destinations would have been the water and grazing resources along the upper Niger River valley and its tributaries, the Senegal River valley, and Lake Chad, then much larger than it is now. The Lake Chad area seems to have been occupied by yet other Nilosaharans, in all probability growing sorghums. In the other two areas, however, the herders would have encountered members of the Mande and Atlantic branches of the Niger-Congo language group. It was among them that agricultural systems of production were developing around African rice, fonio, and guinea millet. From all appearances, the Nilosaharans were absorbed, but they did leave livestock as one of their legacies.

Eventually, pearl millet and sorghums spread throughout the Sahelian and Sudanic zones, cultivated either in the flood plain style known as crue and decrue or else on the interfluves as parts of land-extensive systems of shifting cultivation (Harris 1976). Cattle, sheep, and goats also spread throughout the region, although often herded by specialists, most notably the Fulbe or Fulani. Their origins are traceable to the Senegal River valley, where they may have arisen out of a synthesis of Nilosaharan herders and resident Atlantic-speaking cultivators (Smith 1992).

The Kintampo sites demonstrate that agriculture was on the fringes of the tropical forest zone of western Africa more than 3,000 years ago. Crops were probably still less important than gathered plant foods, but there is strong evidence that goats and sheep had achieved considerable economic importance (Stahl 1993). How long it took from the time of Kintampo for agriculture to reach into the forest itself cannot be determined, although it is unlikely that much of a move was made prior to 2,000 years ago. Then it arrived from two directions. In the west it accompanied the migrations of Atlantic- and Mande-speaking peoples who brought their African rice with them. Farther east, yams accompanied the expansion of other Niger-Congo speakers, this time affiliated with the Kwa and Ijoid families. Eventually, a line that still holds today at approximately the Bandama River in the Ivory Coast came to separate distinctive rice-and yam-eating zones.

A few domesticates from the tropical forest were added to the crop inventory as supplements to the staples. Those in most widespread use include okra (Hibiscus esculentus), kola (Cola sps.), akee (Blighia sapida), and the melegueta pepper (Aframomum melegueta). Also associated with the forest are dwarf varieties of humpless short-horn cattle that are somewhat tolerant of virtually ubiquitous sleeping sickness but yield very little milk. Most likely, they are adaptations of the types represented at Tassili (Blench 1993).

Ethiopia and the Horn

In general, the Ethiopian food crop domesticates remained rather narrowly confined to their original environments. The major exception is finger millet, which spread along with the migrations of Southern Cushites that began around 4,000 years ago. A branch of the Cushitic division of the Afroasiatic group that originated in Ethiopia, these people left their highland home for present-day Kenya and Tanzania (Ehret 1979). Taking with them livestock as well, the Southern Cushites were entering virgin agricultural country, but in such few numbers that most of it remained the province of hunter-gatherers until the arrival of later agriculturists.

Some other Cushites had become livestock specialists and claimed the semidesert and desert lands surrounding the Ethiopian highlands. The most ancient are the Beja (the Blemmeys of the Egyptians), who took possession of the lands astride the Red Sea coast. When their migrations first began is unclear, but they probably date to between 4,000 and 3,000 years ago. Camels were crucial to their advance through the desert, and if the time estimates are correct, they would have been Africa’s original camel herders and the likely source for the animal’s introduction into Egypt. Later, camels fueled the expansions of the Afar and Somali, who occupied the arid lands stretching from Djibouti to the Tana River in Kenya.

Equatorial Africa

The first agriculturists to approach the equatorial rain forest were proto-Central Sudanic (Nilosaharan group) speakers. They did not, however, enter the forest, remaining instead in the savanna and woodland country to the north, where they could cultivate sorghums and tend herds of cattle, sheep, and goats. Getting closer were Ubangian communities whose settlements reached the northern fringes of the forest (David 1982). Early yam-oil palm cultivators located in the Cameroon highlands, they had begun an expansion around 5,000 years ago that took them eastward through the lands straddling the watershed of the Nile and Congo rivers.

It was left to yet another group of yam-cultivating Niger-Congo speakers, the Bantu, to take food-producing economies into the rain forest proper. Originating in the vicinity of the Cross River valley of present-day Nigeria, they had entered the equatorial forest by 4,000 years ago (Ehret 1982).Avid fisher folk as well as farmers, the pioneers followed the Atlantic coast and river valleys west of the immense swamp along the lower reaches of the Ubangi River before it joins the Congo. Subsistence needs kept the frontier moving until advance communities eventually reached the forest’s southern end. Some people then headed east and followed northward-flowing rivers back into the forest, completing a cycle by 2,000 years ago that created a linearlike pattern of Bantu-speaking fishing-farming villages surrounded by bands of pygmoid gatherer-hunters in the deeper forest recesses (Vansina 1990).

Eastern Africa

The transition of eastern Africa from a region of gatherer-hunters to agriculturists was linked to a complex interweaving of population migrations that spanned some 3,500 or more years. As noted, peoples of Southern Cushitic linguistic background introduced finger millet-based agriculture into parts of eastern Africa. Their migrations seem to have centered on better-watered upland areas, such as the slopes of mounts Kenya and Kilimanjaro (Nurse 1982).

At about the same time, or only slightly later, Central Sudanic agriculturists with sorghum and livestock entered the interlacustrine region of present-day Uganda, and may have eventually reached as far south as the Zambezi River valley. As with the Southern Cushites, their numbers were few, and they gave way between 3,000 and 2,000 years ago to Bantu migrants who had tracked through the equatorial forest following easterly rather than southerly routes. Adding Central Sudanic grain-farming and livestock-rearing technologies to their yam-based agriculture, the Bantu were set to claim most of eastern Africa’s cultivable lands as their own (Ehret 1982). By 500 years ago, they had created a wide variety of agricultural systems, ranging from intensively irrigated ones to those using land-extensive methods of shifting cultivation.

The noncultivable lands of eastern Africa, ranging from semiarid to desert, fell to various herding peoples. The north of what is now Kenya was claimed by Cushitic-speaking Somali and Oromo, who (because of the aridity) emphasized camels, while cattle-oriented Nilotes (Nilosaharan group) took possession of the richer grassland and savanna habitats of the interior plateaus.

Southern Africa

As in equatorial and eastern Africa, food-producing economies reached the southern portions of the continent in the company of migrants. Some of them were Khoe-speakers, now called Khoikhoi (formerly Hottentot). They acquired (probably from Bantu sources) sheep and maybe goats, but not crops, somewhere in the vicinity of the middle portions of the Zambezi River valley about 2,000 years ago. From here they spread southward into the semiarid range lands bordering the Kalahari, and thence into the summer rainfall country from the Orange River valley to the Cape of Good Hope, either displacing or intermingling with bands of San (formerly Bushman)-speaking hunter-gatherers (Elphick 1977).

Following the Khoikhoi, but entering different environmental zones, were various communities of Bantu. Some were related to those who had moved through the equatorial rain forest. Continuing on a southward course, they entered the vast expanses of drought-prone and soil-poor brachystegia or miombo woodlands that stretch beyond the Zambezi River. Poorly suited to yam cultivation, at the outset the migrants were forced to rely on gathering, hunting, and fishing for subsistence. But then, sometime just prior to 1,500 years ago, sorghums and millets were acquired from yet other Bantu groups that had begun moving south from eastern Africa. With these crops, much of the area could support at least land-extensive forms of shifting cultivation. Although livestock also were introduced, their roles were restricted by the widespread prevalence of sleeping sickness.

Much better habitats were encountered by the eastern stream of Bantu. One favored route led around lakes Tanganyika and Nyasa and onto the Zimbabwe plateau, and thence to the high veld of South Africa. By A.D. 500, these areas, free from sleeping sickness and comparatively well watered, had been reached by advance communities with grain-growing and livestock-herding economies, although cattle do not seem to have been important until the advent of the New Iron Age after A.D. 1,000. Yet another route followed by Bantu agriculturalists went along the coast through Mozambique and into Natal, which also seems to have been reached by around 1,500 years ago (Hall 1990).

It thus took about 6,000 years from the signs of its first appearance for agriculture to make its way to the extreme southern end of the continent. To an almost exclusive degree, the crops involved were of African origin. However, some of the more important later developments were hinged to imports, and it is to these that the discussion now turns.

Later Imports

Plantains and Bananas

The first of the later imports to have widespread agricultural repercussions in Africa south from the Sahara were plantains and bananas (Musa sp.). Two routes of entry from their southeast Asian source of origin seem possible. The first could have brought them to Egypt over 2,000 years ago via trade with India; from Egypt they spread along the Nile to its headwaters in the Great Lakes region of eastern Africa. The second possibility involves Malayo-Polynesian voyages to the east coast on the way to settling Madagascar. Cuttings could have survived the long sea voyages, and therefore joined other crops, such as cocoyam (Colocasia antiquorum) and sugarcane (Saccharum officinarum), sometime during the second half of the first millennium A.D.

By whatever route, the adoption of bananas and plantains became crucial to the spread of agriculture through the lowland equatorial rain forest. People now had a high-yielding crop that allowed for relatively permanent villages to be established away from the river valleys and deeper into the forest proper. Bananas and plantains also played an important role in augmenting agricultural productivity throughout the moister zones of the Great Lakes region and in the uplands of eastern Africa, such as around the lower slopes of mounts Kenya and Kilimanjaro, thus helping to create some of Africa’s most densely settled areas outside Egypt.

Asiatic Yams

It is possible that Asiatic yams (principally D. alata and D. esculenta) also reached Africa with the Malayo-Polynesian voyages. Their rise to prominence, however, seems mainly to have followed their importation by the Portuguese during the sixteenth century. Because of higher yields, they have largely replaced the indigenously developed varieties throughout the yam belt of western Africa and elsewhere, to the point where the latter serve more for ritual than staple-food purposes (Coursey 1976; Harris 1976).


Reports continue to circulate about the cultivation of maize having been widespread prior to the Columbian voyages, but it was only after the introduction of American varieties (Zea maize) that the crop became an important component of many African food systems. Although the evidence is far from definitive, it does point to the Portuguese as, once again, the carriers, and if this is true, then maize probably had many points of entry along the Atlantic and Indian Ocean coasts. Indeed, there is really no other way to explain its seemingly sudden appearance by the seventeenth century in places as far apart as the Senegal River valley, the Congo basin, the Natal coast of South Africa, and Zanzibar (Miracle 1966).

The frontiers of maize have been moving ever since, to the point where today its now numerous varieties have become the predominant staple for many people in Africa south from the Sahara. Areas where it is particularly important include the highlands of Kenya, Rwanda and Burundi, Malawi, the coastal savanna gap from eastern Ghana to Benin, the Bie Plateau of Angola, and the well-watered areas from Zambia to South Africa. The attractions of maize have been its high yields and invulnerability to bird attacks in comparison with millets and sorghums. About the only places where it does not show up among the list of important crops are in the lowland equatorial zone, the Sahel, and highland Ethiopia, and among rice growers in western Africa. Maize is Africa’s most researched food crop, and the development of varieties that are more drought resistant and that will grow in cooler highland climatic conditions has added to its attractions.


There is little doubt about cassava (Manihot utilissima or M. esculenta) having been brought to Africa by the Portuguese. They found it growing in Brazil and, as with maize, seem to have taken it with them to many of their African landfalls. Initially, the most important of these was the Kongo Kingdom, located south from the Congo River below Malebo Pool. Cassava gained favor here early in the sixteenth century and rather quickly spread both north and south around the equatorial rain forest, apparently being welcomed as a high-yielding alternative to yams and sorghums (Jones 1959). As the slave trade intensified across this region, cassava became even more important. Harvesting could await the disappearance of the turbulence brought by slavers, and cassava flour could be made into cakes for storage and long-distance travel. As the demand grew, some peoples responded by forming their own slave-run plantations to grow cassava. Ironically, its appearance in central Africa probably kept populations growing sufficiently so that the slave trade and slavery could continue (Miller 1989).

A link to slavery may also partially explain cassava’s importance in and around the Niger delta, although much of its spread there, and elsewhere in coastal West Africa, seems to have occurred after 1800 (Jones 1959). Commercial agriculture had begun to expand across the region from Ghana to Nigeria, and cassava’s adoption may have been a response to a decline in other food crops as more land was given over to cacao (Theobroma cacao), another import from the Americas, and oil palm cultivation.

Cassava’s tolerance to a wide range of soil types and moisture conditions, along with the ability to remain unharvested after ripening, made it a favorite of many colonial agricultural officials. They viewed cassava as an ideal famine-prevention food, and thus encouraged plantings in such areas as the upper Guinea Coast and interior eastern Africa where it had not found an earlier enthusiastic reception. Today cassava is tropical Africa’s most widely planted crop, and the region has become the world’s leading producer.

Two other American crop introductions need to mentioned. One is the groundnut (Arachis hypogaea). Groundnuts serve both as a food and cash crop for many Africans in the subhumid and semiarid parts of the continent, with the rural economies of Senegal and northern Nigeria particularly dependent on them. The second crop introduced from the Americas is the sweet potato (Ipomoea batatas). Although not really a staple anywhere in Africa, it is nevertheless an important secondary food source virtually everywhere it can be grown.

Food Preparation and Eating Habits

In Africa a staple food crop is just that – the principal item in the diet, eaten most days if not virtually every day. If a grain is used, most often it will be pounded into a flour, whereas root crops frequently are grated before being boiled. In either case, the end product is a stiff porridgelike substance that is normally formed into bite-size balls. These are then eaten with a variety of stews and soups. Greens of one kind or another will almost always be one of their ingredients, and onions are increasingly common. Otherwise, the additives will depend on local availability and preferences. Meat is always desired but is infrequent. Along coasts and rivers, small amounts of fish are commonly put into stews and soups, and oils, mostly from palms and groundnuts, are in regular use. Salt is a necessity, but the use of other spices varies considerably. In general, tastes tend to be on the bland side throughout most of eastern, central, and southern Africa, while in parts of western Africa and Ethiopia, the stews are often fired with chilli peppers.

Other kinds of preparations include roasting maize ears and steaming sliced cassava, plantains, and sweet potatoes in leaf wrappers. Rice is prepared in the usual boiled manner, and to an ever-increasing extent, imported varieties are preferred because of their higher quality. Such preferences are especially true of urban dwellers, who have created a demand for imported wheat flours to make leavened breads as well. Cassava flour is also used for bread, and in Ethiopia a flat bread made from teff or wheat is the staple source of carbohydrates for people living in the highlands.

All kinds of other foods, especially gathered ones, find their ways into African diets. By far the most common are greens of one kind or another; mushrooms are also widely eaten. Termites are popular, as are various caterpillars and mollusks. These days, game meats are increasingly uncommon, and the vast majority of people will never have occasion to eat them. In fact, game meats have fallen out of favor, even when made available as a part of government game-cropping schemes. Far preferred, if affordable, are beef, mutton, and goat. Vegetarianism by choice is not part of any African cultural tradition.

Contrary to what might be expected, pastoralists, such as the Fulbe and Maasai, also eat very little meat. They rely on fresh and soured milk, along with butter, as their staples, and slaughter animals only for special ceremonial occasions or when these become old and feeble. Milk products are uncommon for most other Africans, as is indicated by the high rates of lactose intolerance that exist, especially in western and equatorial Africa (Kretchmer 1977).

The normal pattern is for one large meal to be served each day. Both morning and evening meals tend to be more like snacks and often involve leftovers. Almost everywhere, men eat separately from women and young children and are served first. Individual plates are not used; rather, each person takes from common bowls and pots.

Meals are seldom accompanied by beverages, except if these are part of feasts. The drinking of alcoholic beverages, however, is virtually ubiquitous, except among Muslims. Traditional products are all basically variations of two themes. The most widespread is the fermentation of one of the crops into something beerlike. Finger millet is the grain generally employed by many eastern and central Africans, whereas maize is popular in South Africa. Also widely used are sorghums and bananas, with the higher sugars from the latter yielding greater alcohol content. A favored practice is to add honey to the fermentation to push the alcohol level even higher. The second theme is found mainly within the equatorial zone and along the coast. Here palm trees are topped to produce a sap that is turned into palm wine, which achieves greater strength than the beers.

In recent decades, commercially distilled spirits and brewed beers have become the alcoholic drinks of choice for many people. As a consequence, alcoholism is now a serious social and health problem, especially in Africa’s rapidly growing cities.

Many food avoidances exist, the most widespread of which is attached to eggs. Some peoples, particularly in Kenya, eschew eggs altogether, equating them with excrement, but more often the restriction falls on girls and reproductive-age women. Reasons stated vary from eggs causing sterility or excessive sexuality to adversely affecting fetal development (Simoons 1961). Less widespread, but of considerable cultural and historical interest, is the avoidance of fish by peoples from Ethiopia southward through Kenya to Tanzania. They are all of Cushitic linguistic affiliation or others who have been heavily influenced by them (Murdock 1959).The reasons are lost in antiquity, but may relate to the low-caste status of fisher folk in the development of Cushitic society, one in which the vast majority of people are either farmers or herders.

Diet-Related Disorders

As a whole, Africa south from the Sahara has become characterized by a worsening food-supply crisis. This is illustrated by the sequence of famines, beginning with the Sahel and Ethiopia in the early 1970s and continuing to Somalia and Mozambique at the outset of the 1990s, and by a regionwide decline in per capita food production over roughly the same period of time. Only by the importation of more and more staple foodstuffs have famines and near famines been averted elsewhere.

Both because of their episodic nature and the coverage they have received, Africa’s famines are not discussed here beyond a note of the complexity of their causes. Although natural events, particularly droughts, have often been involved, these have largely been triggers setting off deeper economic and political charges that initiate the famines. Attention is focused here, however, on more chronic diet-related disorders. These include the various syndromes of protein- energy malnutrition (PEM), vitamin A deficiency, anemias, and goiter.

Protein-Energy Malnutrition

There is very little doubt that, taken together, the syndromes of protein-energy malnutrition constitute the most widespread and serious diet-related disorders in Africa south from the Sahara. From the sketchy evidence available, it does seem that pure cases of kwashiorkor are rather rare today. Apparently, all of the earlier research on it has paid off, particularly with regard to recognizing kwashiorkor in its early stages and alerting parents to take children with symptoms to clinics and hospitals (Newman 1995). The current status of the other two clinical syndromes of marasmus and marasmic kwashiorkor is less certain. Early weaning onto inadequate breast-milk substitutes and exposure to bacterial contamination are still widespread, and in fact, the kinds of poverty, especially in urban areas, that favor this combination of health threats have been growing rather than declining. Unfortunately, the available medical and survey data are inadequate to the task of making any kind of reasonable statements about comparative frequencies.

Somewhat more can be said about the several growth measurements that are used to identify mild-to-moderate forms of PEM. Stunting, for example, is widespread, with rates for children under 5 years of age running from a low 16.6 percent in the Ivory Coast, to 47.8 percent in Burundi, to 53.7 percent in Malawi (Carlson and Wardlaw 1990). One can argue about the standards used for such determinations, but what we know from other sources about food availability in such places as Burundi and Malawi would confirm that many young children are suffering from chronic malnutrition that quickly can go from mild-to-moderate to clinical when infections strike.

Vitamin A Deficiency

Vitamin A deficiencies that lead to vision impairment are far more localized than PEM. They are almost completely absent from humid tropical regions, where sources of preformed vitamin A and beta-carotene are abundant. In contrast, these can be in short supply (especially seasonally) in drier environments, and this is where the problems are concentrated. Some fairly good studies of different time periods have come from Nigeria, and they document a north-to-south incidence gradient (Nicol 1949; Oke 1972). Similarly, the dry Luapula Valley is Zambia’s trouble spot, with a recent study having recorded a deficiency rate of 75 percent among preschool children (Lubinga 1989).

The symptoms of severe vitamin A deficiency run a gamut from Bitot’s spots, through night blindness, to keratomalacia. When the latter stage is reached, death usually follows within a short time because of the problems of caring for the blind in physically hazardous environments. Treatment of vitamin A deficiency is inexpensive and can easily prevent the occurrence of impaired vision, but sadly, access to treatment is still wanting in poorer, more remote areas (Sommer 1982).


Two kinds of diet-related anemias seem to afflict many African women and are especially health threatening during their reproductive years. One is the megaloblastic variety, which develops from a severe deficiency of folic acid. The primary source for folic acid is green leafy vegetables, and most women do, in fact, ingest adequate quantities of these. However, seasonal shortages can occur, especially in more arid habitats, and during times of drought, greens may be almost totally unavailable.

A more widespread anemia stems from iron deficiency, although seldom is the cause dietary. Sometimes it results from metabolic malfunctioning that inhibits iron absorption, but much more common are cases associated with parasitic infestations, notably malaria, hookworm, and roundworm (Latham 1965).


Goiters, resulting from a dietary deficiency of iodine, can be encountered in many parts of the continent where the ancient basement geological complex predominates as the bedrock. Because it lacks iodine, so, too, do the soils and, therefore, the plants. The center of endemicity seems to be Zaire, where incidences of up to 100 percent have been recorded within some communities (DeVisscher et al. 1961). Also possibly contributing to the high rates are goitrogenous agents in widely eaten cassava and cabbage. The majority of goiters occur to young women because of the thyroid stresses resulting from menstruation, pregnancy, and lactation. Most are temporary, having little or no health impact, although several studies have found a connection between community incidences of endemic goiter and cretinism (Bastinie et al. 1962; Delange and Ermans 1971).