West African Late Stone Age

Dr. Susan Keech McIntosh and Sarah Berry

Later Stone Age (LSA), Ceramic Late Stone Age

30,000-4000 BP

Follows the Sub-Saharan Africa Middle Stone Age Tradition and precedes the West African Neolithic Tradition. Microlithic assemblages are present in Gabon from 40,000 BP and at Shum Laka from 30,000 BP. Areas with hunter-gatherers using flaked stone tools may persist well after 4000 BP in some areas, particularly the southwest. The transition to the Neolithic—marked by domesticated animals and/or plants in the diet—is well-documented only in the Sahel and northern savanna, and in central Ghana.

The West African Late Stone Age is found throughout sub-Saharan West Africa, in a variety of different vegetation zones and environmental contexts. The western and southern borders are the Atlantic Ocean and Gulf of Guinea. The northern border is the Sahara desert north of the Niger River. The eastern border extends from Mount Cameroon to Lake Chad. The area covered by the tradition would have changed with variations in climate during the Late Pleistocene and Early Holocene, with wetter periods expanding the Sahel and the viable area northward, into what is today the Sahara Desert.

A diagnostic trait is the presence of microliths. Microliths are defined as small tools that often have one or more sharp edges, with at least one edge trimmed by steep blunting techniques. Some areas also have large flake-tool technologies. Ceramics appear late in the tradition.

Too few sites from this period have been adequately excavated and published to clearly determine subtraditions, but alongside characteristic microlith assemblages the West African Late Stone Age includes larger flaked tool assemblages, previously known as Tumbian, Para-Tumbian, Guinea Neolithic, Neolithic hoe cultures, and Sangoan.

Bosumpra Cave, Ghana; Fanfannyégèné, Mali; Iwo Eleru (Ihò Eleru), Nigeria; Kamabai, Sierra Leone; Kourounkourokale (Korounkorokalé), Mali; Manianbougou (Manianbugu), Mali; Mejiro cave, Nigeria; Rim, Burkina Faso; Rop, Nigeria; Shum Laka, Bamenda, Cameroon; Vallée du Serpent sites, Mali; Yengema, Sierra Leone

In the Terminal Pleistocene, until 12,500 BP, the climate was much colder and drier than at present, with the desert margin shifted 500 km south of its present location and the forest reduced to remnants in the Niger Delta, Cameroon, the Ivory Coast/southwestern Ghana, and southeastern Liberia. Early microlithic assemblages—the geometric assemblage at Iwo Eleru (11,000 BP) and the non-geometric assemblage at Bingerville Highway in the Ivory Coast (13,000 BP) —date to this period. By 10,000 BP the climate became dramatically wetter, with a favorable precipitation/evaporation balance assured by cooler temperatures. This change was due in part to a shift in the Intertropical Convergence Zone, a belt of low pressure that moved about 500 km northwards, bringing the West African Monsoon and more rainfall during the early and middle Holocene. The enhanced rainfall created lakes, rivers and inland deltas in what is presently the Sahara. Forest recolonization occurred in Ghana, and Paleolake Megachad formed at the 40+ m level of present-day Lake Chad; its levels remained high until 8000 BP and then declined, with a rapid fall during an arid interval between 7500-7000 BP. The evidence for a dry episode between 8000–7000 BP is not universal across West Africa. In the Manga Grassfields of northern Nigeria, for example, pollen cores from four sites show no discernible arid event at that date, despite proximity to Lake Chad. Lakes in the Malian Lake Region of the Niger River did not all evaporate and may have provided usable habitat for humans. Precipitation in coastal Ghana may have decreased by half. The second Holocene humid period dates to 7000–4000 BP based on evidence in the Malian Lakes Region. There is a growing interest in identifying synchronous climatic events in Africa and beyond that may correlate significantly with major shifts in subsistence and technology. But the considerable asynchronicity in pollen profiles at sites less than 100 km apart in northern Nigeria reminds us that local factors such as water table depth, topography and fluvial influences can alter the impact of widespread changes in rainfall.

After 7000 BP, rainfall became more seasonal, temperatures warmed, and arid episodes of variable intensity and duration increased in frequency. Reconstructions of climate mainly rely on studies of water levels and pollen cores at Lake Chad and Lake Bosumtwi in Ghana, and from Bal Lake playa sediments in northeastern Nigeria. A general trend toward more arid conditions became pronounced after 4500 BP. As a result, streams that once flowed year-round in the southern Sahara became increasingly seasonal and temporary.

The landscape of sub-Saharan West Africa is variable, though with generally low relief; only the Guinean and Cameroonian highlands and the Jos Plateau exceed 1000 m. The Cameroon highlands are volcanic in origin and the soils are fertile. The continental shelf is under 10 miles wide off much of south coast of West Africa; it reaches its widest, up to and exceeding 75 miles, from south of Dakar, Senegal to Sherbro Island, Sierra Leone. Sea level was lower by as much as 125 meters during the Last Glacial Maximum (ca 26,000-19,000 BP), during which there would have been coastal plains available for exploiting additional resources but the evidence for such use is now submerged. The Sahara Desert and its dune fields, along with the Sahel and savanna, expanded and contracted depending on rainfall. The Niger River and its delta expanded with the greater rains of the African humid period of the terminal Pleistocene and early Holocene, and shrank as the rains subsequently diminished. Paleolake Megachad also expanded and then contracted during the Holocene (shrinking to Lake Chad after 2000 BP).

The Precambrian craton of West Africa is widely buried by younger sedimentary rocks and spreads of blown sand, but outcrops occur in the dissected plateaus of the Guinean highlands, the Cameroon highlands and Jos in Nigeria. Sedimentary rock of the Continental Intercalaire and the Continental Terminal strata are present across a wide swath of the western and central regions of West Africa. Quaternary basins and dunes are associated with the Niger/Senegal/Lake Chad catchments. The Cameroon highlands include a chain of volcanoes known as the Cameroon line, extending from islands in the Gulf of Guinea to Lake Chad. One active volcano, Mount Cameroon, is high enough to be dusted with snow on occasion.

Biota varies from Sahelian species adapted to low precipitation (250-400 mm annually with marked seasonality) in the north to tropical forest species (>1400 mm rainfall annually, with little seasonality) in the south. Drier, open Sudanic savanna and more heavily-wooded Guinea savanna occupy the intervening middle belt. The West African landscape of 12,000–7500 BP was very different from today, with interdunal lakes widespread throughout the present Sahelian zone. At that time, Sahelian and Sudanic species ranged north into the present-day central Sahara, and Sudanic and Guinean species were common in the current Sahelian zone.

During the Terminal Pleistocene and early Holocene African humid period, various species of animals were found farther north than they are today. Faunal remains in Saharan archaeological sites include: fish (Perciidae and Siluridae), crocodiles, large bovids, elephants, hippopotamus, and rhinoceros. Ostrich likely were present, as eggshell beads have been found.

During the Last Glacial Maximum, from about 25,000 to 12,000 BP, the climate was colder and more arid, and the Sahara’s tropical extreme desert expanded. Mangrove swamp would have been found along the littoral, and where rainforest is currently located there would have been savanna-grassland. Rainforests retreated into refugia, including one in the highlands of Cameroon and Nigeria. They gradually expanded outward from such areas during the wetter and warmer climate of the Holocene from 11,000 BP to 4,000 BP, and the current vegetation zones (extreme desert, semi-desert, grasslands, savanna, and tropical rainforest) would have been found as much as 400-500 km farther north at the peak of the African humid period. Riverine and lacustrine vegetation, including swamp forest vegetation, occurred along the various lakes, rivers and inland deltas in what is now the Sahara.

Known sites are primarily in caves and rock shelters, which served as temporary camps for mobile hunter-gatherers; Iwo Eleru in Nigeria is a prime example. An open-air site in the Vallée du Serpent, Mali contained rings of cobbles that may be remains of structures.

Human remains have been recovered from Iwo Eleru (one skeleton dated 11,200±200 BP). Pathologies were present, mostly dental, such as abscesses (probably due to pulp exposure in excessively worn teeth).

The full range of plants and animals exploited in the Late Stone Age is not yet known, due to poor preservation and recovery at many of the excavated sites. The present evidence indicates hunting/foraging economies were widespread between 13,000 and 4000 BP. Only Shum Laka, in a wooded savanna/forest environment, has provided a clear picture of a Late Stone Age subsistence economy. Two major dietary components obtained from hunting were the giant forest hog (Hylochoerus meinertzhageni) and the dwarf or forest buffalo (Syncerus caffer nanus). Among the animals hunted were chimpanzee, gorilla, baboon and Cercopithecus monkeys, giant rat, cane rat, and several species of antelope. The giant African snail (Achatina spp.) was also exploited. After the appearance of pottery by 6000 BP, the hunting/foraging economy continued virtually unchanged. New plant resources present at several sites along the forest/savanna margin after 5000 BP include tree products such as the oily seeds of Canarium schweinfurthii and nuts of the oil palm (Elaeis guineensis)—the former utilized since 10,000 BP and the latter since 8400 BP at Bosumpra in Ghana. Historically, oil-palm and yam cultivation have been practiced together; whether yams were used or cultivated at such an early date is unknown. Shell middens attest to littoral resource exploitation in coastal Ghana between 6000-4000 BP. Shell middens began to accumulate near the present shoreline of the Ivory Coast about 3500 BP. Fish and shellfish were also exploited along waterways that existed in the Sahel, the Sahara, and around Lake Chad during the African humid period of the early Holocence. The site of Kourounkorokale, near the Niger River in Mali, contained catfish (Clarias spp.), Nile perch (Lates niloticus), tilapia, crocodile (Crocodylus niloticus), freshwater terrapin (Pelomedusidae), freshwater oyster (Etheria elliptica), a large bovid, warthog (Phacochoerus aethiopicus), lion (Panthera leo), red-flanked duiker (Cephalophus rufilatus), bushbuck (Tragelaphus scriptus), kob ([i]Kobus kob[i]) and several other unspecified species of antelope, and an unidentified equid.

Common stone tools are microliths, defined as small tools that often have one or more sharp edges with at least one edge trimmed by steep blunting techniques, such as snapping or flaking. In areas where good quality stone is available, such as western Senegal, microliths occur in a limited number of standardized, geometric shapes. Elsewhere, where only poor quality raw material is found (often quartz), the assemblage consists of small, unstandardized flakes. Assemblages may have a combination of both standardized and unstandardized forms in widely varying percentages. Typically, microliths are small enough that they were probably hafted for use.

A dugout boat, measuring 8.5 m in length and dated to 8265±275 BP, was excavated at Dufuna in northeastern Nigeria. Bone tools, such as barbed harpoons, have been found at Kourounkorokale. Despite opportunities for mobility, reliance on local raw materials is the rule; all stone at Shum Laka, for example, came from within a 5 km radius. Raw material source sites where stone pick and ax “preforms” were roughed out have been identified in Senegal, Cameroon, Nigeria, and Mali. Previously believed to be late Pleistocene or early Holocene, many of these sites demonstrably date to late in the West African Stone Age or the West African Neolithic.

It should be noted that large flake-tool technologies—in some cases Levallois, in others not—also apparently existed during the early Holocene, although the chronology is not well-established. Both unstandardized microlithic and Levallois flake technologies are present at Iwo Eleru. The reasons for the persistence of larger flake technologies in some areas is unclear; their importance grows after the appearance of pottery. At a number of sites, large core tools such as picks and axes become common. Previously assigned to a large-tool tradition variously known as Tumbian, Para-Tumbian, Guinea Neolithic, Neolithic hoe cultures, and Sangoan, these core tool assemblages cannot, at this point, be separated from microlithic material as representing a different tradition.

Ceramics appear in Late Stone Age assemblages about 6000–5000 BP. The oldest West African site with pottery is Konduga, in northeastern Nigeria, an open-air site on the Bama Ridge dated to 6300 BP. Pottery appears at a variety of rockshelters between the Cameroon Grassfields and Ghana between 5500–5000 BP. Ground stone axes appear at approximately the same time, but flaked stone axes continue to be used. Early on, pottery most frequently has combed or punctate decoration. Simple vessel shapes are common, with rounded bottoms and a high percentage of uninflected rims. All these are elements characteristic of earlier Holocene ceramics in the Sahara.

At Iwo Eleru, 95 percent of the assemblage was made of quartz available within a few kilometers of the site. The remainder was made of chalcedony available 17 km away. Among the 6500 cores identified at Iwo Eleru, over 85 percent were crude and amorphous. Of the 1000 shaped tools recognized, microliths and chisels made up over half the assemblage. Geometric segments and triangles dominated the microliths. At Shum Laka, by contrast, only 75 cores and 28 shaped tools were present (out of over 19,000 pieces of stone). As at Iwo Eleru, the Shum Laka assemblage is also primarily quartz, but the microliths are unstandardized. Unmodified flakes were probably extensively utilized, but identifying use wear on quartz is difficult. [ocm]135 184 324 412 431 911[/ocm} Numerous sites, especially those in the southern half of West Africa, show a pronounced trend to macrolithism following the microlithic industries of the earlier Holocene. The presence of large basalt tools after 7000 BP has been noted at Shum Laka, and large flaked axes appear in the Iwo Eleru sequence at about the same time. Prepared core/Levallois technologies became more common. In coastal Ghana, shell middens are associated with an unstandardized quartz microlithic assemblage. Where geometric microliths are present, their relative frequency varies a great deal from site to site. At present, the reasons for intersite variation in stone tool types and the technologies for producing them is little understood. Differences in raw material quality may account for some of the variability between microlithic assemblages. Interestingly, at Shum Laka (one of the few sites for which detailed studies on raw materials has been done), obsidian, though present within five km, does not increase in frequency through millennia of occupation; nor is it used differently to produce more standardized or more efficient tools with less waste. There is little evidence anywhere in West Africa for extralocal exchange networks for stone until after 4000 BP.

Ostrich eggshell beads have been found in sites in northern Mali.

This tradition summary is based on the article "West African Late Stone Age," by Susan Keech McIntosh in the Encyclopedia of Prehistory, Volume 1, Africa, Peter N. Peregrine and Melvin Ember, eds. New York: Plenum Publishing Corporation, 2001. Sarah Berry expanded a number of sections (Absolute Time Period; Location; Climate; Topography; Geology; Fauna; Flora; Settlement; Subsistence; Ornaments; Bibliography) and made minor revisions to others in January 2021. We thank Peter N. Peregrine for bibliographic suggestions.

Cornelissen, Els (1996). "Shum Laka (Cameroon): late Pleistocene and Early Holocene deposits." In Aspects of African Archaeology: Papers from the 10th Congress of the PanAfrican Association for Prehistory and Related Studies, edited by G. Pwiti and R. Soper. Harare: University of Zimbabwe Publications, 256–263.

MacDonald, K. C. and P. Allsworth-Jones (1994). "A Reconsideration of the West African Macrolithic Conundrum: New Factory Sites and an Associated Settlement in the Vallée du Serpent, Mali." The African Archaeological Review 12: 73–104.

Oslisly, Richard, Hughes Doutrelepont, Michel Fontugne, Hubert Forestier, Pierre Giresse, Christine Hatte, and Lee White (2006). "Premiers Résultats Pluridisciplinaires d'une Stratigraphie Vieille de Plus de 40000 Ans du Site de Maboué 5 dans la Réserve de la Lopé au Gabon." In Congrès de l'UISPP.Union Internationale des Sciences Préhistoriques et Protohistoriques : Session 15. Préhistoire en Afrique = African Prehistory, 14., Liège (BEL), 2001/09/02-08, edited by P. de Maret, E. Cornelissen, and I. Ribot. Oxford: Archaeopress, 189-198. Accessed March 24, 2021. https://horizon.documentation.ird.fr/exl-doc/pleins_textes/divers17-01/010059677.pdf

Cornelissen, Els (2003). "On Microlithic Quartz Industries at the End of the Pleistocene in Central Africa: The Evidence from Shum Laka (NW Cameroon)." The African Archaeological Review 20(1):1-24. Accessed September 2020. https://www.jstor.org/stable/25130759.

MacDonald, Kevin C. (1997). "Korounkorokalé Revisited: The Pays Mandé and the West African Microlithic Technocomplex." The African Archaeological Review 14(3): 161-200. Accessed October 2020. https://www.jstor.org/stable/25130625.

Oslisly, Richard, Lee White, Ilham Bentaleb, Charly Favier, Michel Fontugne, Jean-François Gillet, and David Sebag (2013). “Climatic and Cultural Changes in the West Congo Basin Forests Over the Past 5000 Years.” Philosophical Transactions of the Royal Society B. 368(1625):20120304. Accessed October 20, 2020. https://doi.org/10.1098/rstb.2012.0304.

Coutros, Peter R., and Peter M. J. Douglas (2015). "Coring Lake Fati and Settlement Archaeology of the Middle Niger Lakes Region." The African Archaeological Review 32(2): 249-266. Accessed July 28, 2020. www.jstor.org/stable/43916736.

MacDonald, Kevin C. (1997). "The Late Stone Age and Neolithic Cultures of West Africa and the Sahara." Encyclopedia of Precolonial Africa, edited by J. Vogel. Walnut Creek, CA: AltaMira Press, 394-398.

Petit-Maire, N., J. C. Celles, D. Commelin, G. Delibrias and M. Raimbault (1983). "The Sahara in Northern Mali: Man and His Environment between 10,000 and 3500 Years BP. (Preliminary Results)." The African Archaeological Review, 1: 105-125. Accessed February 10, 2021. https://www.jstor.org/stable/25130427.

Andah, B. (1978). "Exacavations at Rim, Upper Volta." West African Journal of Archaeology 8: 75–138.

de Maret, Pierre, Bernard Clist, and Wim Van Neer (1987). "Résultats des Premières Fouilles dans les Abris de Shum Laka et d'Abeke au Nord-Ouest de Cameroun." L'Anthropologie 91(2): 559-584.

Malhi, Yadvinder, Stephen Adu-Bredu, Rebecca A. Asare, Simon L. Lewis, and Philippe Mayaux. (2013) "African Rainforests: Past, Present and Future." Philosophical Transactions: Biological Sciences 368(1625):1-10. Accessed March 30, 2021. http://www.jstor.org/stable/42569192 .

Phillipson, D.W. (2005). "Chapter 4; Regional Diversification and Specialization." African Archaeology, 3rd ed. Cambridge University Press, Cambridge, 91-146.

Atherton, John (1972). "Excavations at Kamabai and Yagala Rock Shelters, Sierra Leone." West African Journal of Archaeology 2: 39–74

Eyo, E. (1972). “Rop Rock Shelter Excavations 1964." West African Journal of Archaeology 2: 13-16.

McIntosh, Susan Keech and Roderick J. McIntosh (1983). "Current Directions in West African Prehistory." Annual Review of Anthropology 12: 215–258.

Raimbault, M. (1983). "Les Ateliers Préhistoriques de Manianbugu (Bamako, Mali)." Bulletin de l'Institut Fondamental d'Afrique Noire (Series B) 45:219–276.

Breunig, P. (1996). "The 8000-year-old Dugout Canoe from Dufuna (NE Nigeria)." In Aspects of African Archaeology: Papers from the 10th Congress of the PanAfrican Association for Prehistory and Related Studies, edited by G. Pwiti and R. Soper. Harare: University of Zimbabwe Publications, 461–468.

Grove, A. T. (1978). Africa. Oxford: Oxford University Press.

McIntosh, Susan Keech and Roderick J. McIntosh (1988). "From Stone to Metal: New Perspectives on the Later Prehistory of West Africa." Journal of World Prehistory 2(1): 89–133.

Salzmann, Ulrich. and Martyn Waller (1998). "The Holocene Vegetational History of the Nigerian Sahel Based on Multiple Pollen Profiles." Review of Palaeobotany and Palynology 100(1-2): 39–72.

Brothwell, D. and Thurstan Shaw (1971). "A Late Upper Pleistocene Proto-West African Negro from West Africa." Man 6(2): 221–227.

Grove, A. T. (1993). "Africa's Climate in the Holocene." The Archaeology of Africa: Foods, Metals and Towns, edited by T. Shaw, P. Sinclair, B. Andah, and A. Okpoko, London: Routledge, 32-42.

Moeyersons, Jan J. (1996). "Evolution of Shum Laka Rock Shelter (Western Cameroon) Since Late Stone Age Times." Aspects of African Archaeology: Papers from the 10th Congress of the PanAfrican Association for Prehistory and Related Studies edited by G. Pwiti and R. Soper, Harare: University of Zimbabwe Publications, 245-255.

Shaw, Thurstan (1944). "Report on Excavations Carried Out in the Cave Known as 'Bosumpra' at Abetifi, Kwahu, Gold Coast Colony." Proceedings of the Prehistoric Society 10: 1–67.

Coon, Carleton S. (1968). Yengema Cave Report. Philadelphia: University Museum, University of Pennsylvania.

Hassan, Fekri A. (1997). "Holocene Palaeoclimates of Africa." African Archaeological Review 14(4): 213-230.

Nygaard, Signe E. and Michael R. Talbot (1984). "Stone Age Archaeology and Environment on the Southern Accra Plains, Ghana." Norwegian Archaeological Review 17(1): 19–38.

Shaw, Thurstan (1977). "Hunters, Gatherers and First Farmers in West Africa." In Hunters, Gatherers and First Farmers Beyond Europe: an Archaeological Survey, edited by J. V. S. Megaw. Leicester: Leicester University Press, 69-126.

Huysecom, Eric (1990). Fanfannyégèné I. Weisbaden: F. Steiner Verlag Wiesbaden.

Oas, Sarah E., A. Catherine D'Andrea, and Derek J. Watson (2015). "10,000 Year History of Plant Use at Bosumpra Cave, Ghana." Vegetation History and Archaeobotany 24(5): 635-653. Accessed July 28, 2020. www.jstor.org/stable/43554353.

Shaw, Thurstan (1978-79). "Holocene Adaptations in West Africa: The Late Stone Age." Early Man News 3/4: 51-81.

Lavachery, Philippe (2001). "The Holocene Archaeological Sequence of Shum Laka Rock Shelter (Grassfields, Western Cameroon)." The African Archaeological Review 18(4): 213-247. Accessed July 28, 2020. www.jstor.org/stable/25130728.

Shaw, Thurstan and S. G. H. Daniels (1984). "Excavations at Iwo Eleru, Ondo State, Nigeria." West African Journal of Archaeology 14: 1-269.

Lavachery, Phillippe (1996). "Shum Laka Rock Shelter Late Holocene Deposits: From Stone to Metal (Northwestern Cameroon)." Aspects of African Archaeology: Papers from the 10th Congress of the PanAfrican Association for Prehistory and Related Studies, edited by G. Pwiti and R. Soper. Harare: University of Zimbabwe Publications, 265-274.

Stahl, A. (1993). "Intensification in the West African Late Stone Age: A View from Central Ghana." The Archaeology of Africa: Foods, Metals, and Towns, edited by T. Shaw, P. Sinclair, B. Andah, and A. Okpoko. London: Routledge, 261-273.

Linseele, Veerle (2010). "Did Specialized Pastoralism Develop Differently in Africa than in the Near East? An Example from the West African Sahel." Journal of World Prehistory, 23(2): 43-77. Accessed May 5, 2021. http://www.jstor.org/stable/25801287.

Waller, Martyn P., F. Alayne Street-Perrott, and Hongya Wang (2007). "Holocene Vegetation History of the Sahel: Pollen, Sedimentological and Geochemical Data from Jikariya Lake, North-Eastern Nigeria." Journal of Biogeography 34(9): 1575-1590.

Lipson, Mark et al. (2020). "Ancient West African Foragers in the Context of African Population History." Nature 577: 665-670. Accessed September 8, 2020. https://doi.org/10.1038/s41586-020-1929-1.