Ancient Developments in the Middle Senegal Valley and the Inland Niger Delta
Summary and Keywords
The study of West Africa has contributed to the expansion of comparative arid-lands floodplain prehistory, from both the data collection (cultural and historical) and the theoretical aspects. The neoevolutionary approach that often pictures Africa as a backward continent has been successfully challenged. In the Middle Senegal Valley and in the Inland Niger Delta, research on their societies’ complexity done along these two subcontinent’s floodplains has described new processes (including urbanization) that were not previously featured in the archaeological literature. The two floodplains, because of their ecological diversity, with the richness of their ecological diversity, attracted Saharan populations affected by drought at the end of the second millennium and the first millennium BC. However, after their initiation occupation the two areas took different trajectories in complexity and settlement organization. Large complex settlements have been found at Jenne-jeno and in the Ile a Morphil that illustrate whole new trajectories of civilization. These forms of complexity, found in areas with historically known polities, were not included in the range of possibilities predicted by standard complexity theories regarding civilizational development. Ethnographic and historical data, reveal the existence of societies with a central authority embedded within and balanced by a diffuse, segmented and heterarchical power structure; often as a strategy to resist the individual consolidation of power. These societies exhibit evidence of horizontal differentiation and consensus-based decision making. All these types of organization are characterized by the presence of several sources of power vested in corporate entities, such as lineages, age groups, cults and secret societies.
Floodplains of the world have been classic locations for archaeological investigation of preindustrial urbanism, interregional trade, and emerging social complexity. The arid-land floodplains are good comparative laboratories in which to test theories of emerging social ranking and state formation. They are particularly good locations for theory testing because of their shared characteristics of rich soils, redundancy of other resources (encouraging long distance trade), easily monopolized access to water (encouraging power control), and potentially dense populations (facilitating coercive institutions).
West Africa is contributing in shedding light on processes of complexity in a global perspective. The areas that are most referred are the Inland Niger Delta and the Middle Senegal Valley (Figures 1 and 2).
Located in the Sahel, the Middle Senegal valley and the Inland Niger Delta are the two biggest floodplains in West Africa. The Sahel is a semiarid climate zone located between the Sahara and the subsavannah. It is characterized by its weather pattern (long dry season vs. short rainy season), its specific rainfall patterns (between 100 and 600 mm of rain per year) and its vegetation covering (dominated by the acacia, spiny trees, shrubs, bushes, and grasses) (Figure 3).1
The weather pattern is influenced by two anticyclones: the desiccating North African anticyclone with its warm and dry winds called harmattan blowing northeast-southwest and the anticyclone of Saint Helene with its humid air masses. These two anticyclones are separated during the year by the fluctuating position of the moisture-bearing Inter Tropical Convergence Zone (ITCZ). The relative strength and latitudinal position of the anticyclones determine the location seasonally of the northeast dry air mass (alize) and the moisture-laden southern air mass (monsoon). The northeast dry air mass dominates during the dry season, whereas the monsoon dominates during the rainy season.
Overall, the Sahel is a very sensible climate zone marked by climate fluctuations at the scale of decades, centuries, or millennia.2 This has pushed populations to develop, in the past, several coping mechanism such as modification of subsistence strategies, mobility, and creation of alliances networks.3
Climate and Hydrology
Middle Senegal Valley: The Middle Senegal Valley is a reasonably fertile floodplain located between Senegal and Mauritania. The climate today is characterized by a high interannual variability and major climate fluctuations. The annual rainfall varies between 300 mm and 500 mm. From 1930 to 1960, the annual mean rainfall was 327 mm per year. The weather in the Middle Senegal valley is characterized by a succession throughout the year of a long dry season (nine months) during which the harmattan blows, followed by a short rainy season (three to five months), during which the monsoonal rains prevail.
Using two climate parameters (rainfall and temperature), local populations have divided the year’s climate into five seasons: Dabunde, Ceedu, Deminare, Ndungu and Kawle.
Dabunde goes from December through March. Temperatures are relatively cool during that season. Small rivers and shallow ponds dry out; trees lose their leaves. Small plants dry out (except those that have thorns). Peasants gather millet and sorghum.
Ceedu starts in March. It is the warm season attributed to the excessive heat carried by the harmattan. Temperatures sometimes reach 40° C. That heat contributes to the final withering of plants and rivers. Consequently, soils become bare, and unprotected by vegetation. Winds are very active, causing dust transport and massive deflation. Indeed dust winds carrying soils and sediments from the Sahara penetrate the MSV. They reduce visibility and reduce the possibility of any subsistence activity.
Deminare is the spring. It starts in May and precedes the rainy season. Trees bud. Temperatures are cool during the evening.
Ndungu is the rainy season. It goes from July through October. Peasants grow millet, sorghum, and peanuts. It is also the season of most active fishing. The first rains are very stormy and very destructive for the soils that have lost their vegetative protection. The rain causes big gullies and runoff. The progressive expansion of the flood, combined with the filling up of basins and low-lying surfaces, transform several areas of the valley into large seasonal ponds. However, topographic units such as the high levees and the red dunes are rarely flooded.
Kawle marks the end of the rainy season. It is the harvest season (October through December). It is characterized by the persistence of the humidity (that starts during the ndungu) and very strong solar rays called naange kawle (kawle sun).
The hydrology in the Middle Senegal Valley is dominated primarily by the Senegal River, and secondarily by its distributaries. From its sources at Bafoulabe to the delta, the 1784 km long Senegal River crosses three different climate zones (subforest, Sudano-Sahelan, and Sahelo-Saharan zones) and is fed by a catchment basin 335,000 km2 wide.4 The river originates at Bafoulabe (in the Fouta Jallon Mountains) at the junction of two of its tributaries: the Bafing and the Bakoy. Upstream from Bakel, the Senegal River is strengthened by the added discharge of three other tributaries: the Faleme, the Karakoro and the Kolombine.
The Upper Valley, situated in a wet climate zone, gets more rain than the Middle Valley and the Delta. However, the narrowness of the bank in the Upper Valley, combined with the presence of rocky topography not suitable for agriculture, restricts the advantages local populations can gain from higher precipitation. Downstream of Bakel, the majority of the water pours onto the silty soils of the Middle Valley (situated in a semiarid climate zone). These soils are more suitable for agriculture and for other subsistence activities. Consequently, even though the most important portion of discharge comes from a forest zone, populations who live in the semiarid zone (Middle Valley) are able to take greater advantage of the flood than the people of the Upper Valley.
The flood occurs during the rainy season. The flood’s rise starts in May–June in Guinea. The duration of the flood depends on the length of the rainy season, the nature of the rain, the texture of the soil, and the elevation. The first flood flows reach the Middle Valley in July.5 The expansion of the flood is slow because of two reasons: the shallowness of the slope and the fact that the first flood flows are greedily swallowed by soils that have been desiccated for several months. In July–August, the water fills the larger bank, flooding basins and channels while leaving the high levees and red dunes dry. The channels (called belli or vendou in singular) vary in duration: some are seasonal, whereas a few are permanent.
There are several topographic units defined based the geomorphology, land use, and perception. Each unit has its specific aspect of pedology, grass cover, and tree species. Each unit has its specific utility to past and present populations (Figures 4a and 4b).
Inland Niger Delta: Like the Middle Senegal valley, the climate in the Inland Niger Delta is influenced by the Inter Tropical Convergence Zone (ITCZ) that engenders a long dry season lasting nine months (October–June) and a short rainy season lasting three months (July–September).
The hydrology in the Inland Niger delta is dominated by the Niger River. That river is 4100 km long; ranking it as the third longest African river (after the Nile and the Congo). From its source in the Loma mounds (between Conakry Guinea and Sierra Leone), it moves upwards along a northeast axis, crossing forest, savannah, and semiarid zones. 800 km from its source, the river enters the inland delta: a vast flat floodplain of pseudo-deltaic hydrology.
Located between the latitudes 17° et 13° North and the longitudes 2°30 et 6°30 West, the Inland Niger Delta is 40 000 km2 wide. It includes medieval towns such as Jenne, Mopti, and Tombouctou. It is watered by a hierarchized network of channels dominated by the Niger River and its confluent, the Bani (a river from Ivory Coast that joins the Niger River after Mopti).6 Local people call the Niger River Joliba (Mande) or Issa Beri (Sonrai).
The water from the Niger River and the Bani fills the vast mosaic of swamps, canals, and lakes such as lakes Debo and Korientzé.7 Hence, the hydrology and the mosaic of landforms have contributed to making the Inland Niger Delta the most important water supply in that part of the world, attracting pastoralists, farmers and fishermen who have developed different adaptation strategies.8 In fact, over one million people practice subsistence activities that depend on the Niger River and the Bani.
Because of its location within the Sahel and south of the Sahara, the attraction of the Inland Niger Delta is deeply rooted in the past. According to Roderick McIntosh:
The late Holocene Inland Niger Delta should have proved an attractive magnet to Late Stone Age peoples fleeing the final desiccation of the Sahara.9 Indeed, the earliest pottery at Jenne-jeno shows unmistakable Saharan affinities. The abundant floodplain resources in the survey region would have allowed a continuation of the Holocene Saharan subsistence strategies which were heavily reliant upon fish and wild grass and increasingly upon pastoralism. The environment would have been appropriate for experimentation leading to the domestication of African rice, which has long been argued, on botanical grounds, to have occurred in the middle Niger region.10 (S. K. and R. J. Mclntosh, 1980, Part i)
After Timbuktu, the Niger River describes an immense curve heading east, and after Gao, it heads southeast for approximately 2000 km to finish its course in the Nigerian side of the Gulf of Guinea.
The rain lasts in general three months (July–September). The maximum rainfall occurs in August. In that semiarid environment, much of the subsistence activities depend on the flood which lasts six to nine months (July–November). This discrepancy between the duration of the rainy season and that of the flood is due to the fact that much of the flood in the Inland Niger Delta comes from rainfall upstream (where the annual mean rainfall is 2000 mm). The flood occurs slowly in the Inland Niger because most of the soils stay dry for most of the year. In fact, because of the semiarid conditions and the slowness of the slope, between 25 and 50 percent of the water is lost by evaporation.11
The volume of the flood and the extension of the areas flooded depend on the climate. In time of positive rainfalls, more areas are flooded than in time of drought. This is why the curation of a traditional coping mechanism is very present in the local population’s sociocultural memories.12 Rice is cultivated during the flood; millets and sorghums are cultivated during its recession. Susan McIntosh and Roderick McIntosh cite Cissokho, who stated that the Inland Delta was the major food supplier of the Saharan trade towns 500 to 1000 km away.13
Historical and Archaeological Background
Besides being located in the same climate zone (the Sahel), the Middle Senegal valley and the Inland Niger Delta share strong similarities in historical and archeological backgrounds. Historically they were visited by the same Arab geographers; hence described by the same witnesses. The two areas were also colonized by the same country: France. This has engendered the same colonial historical perspective that was based on colonized history and its corollaries: the stimulus-diffusion paradigm. That paradigm was later supplanted by the internationalization of the research that engendered new data, new perspective, and new findings illuminated by proponents of the processual approach the most successful of whom were Roderick McIntosh and Susan McIntosh in Mali starting in the 1970s and in Senegal starting in the 1990s.
Primary Historical Sources
The earliest historical information on the Middle Senegal Valley and the Inland Niger Delta comes from the account of Arab geographers, scholars, and travelers. They went to West Africa for two related reasons: the trans-Saharan trade and the expansion of Islam. These two transformative events influenced the social, economic, religious, and political fabrics in Middle Senegal Valley and the Inland Niger.
Chronologically, there are two phases of Arabic sources: the first phase expands for five centuries (9th to 14th centuries). The authors were outsiders who had an outside view of West Africa and its people: Ya˓ḳūb (m.ms 897 CE), al-Mas˓ūdī (ms 965 CE), Ibn Ḥawḳal (ms 977 CE), al-Bakrī (ms 1068 CE), al-Umarī (m.s 1342 CE), d’al-Idrīsī (ms 1154 CE), Ibn Sa’īd al-Gharnata (ms around 1288 CE), al-Umarī (ms 1342 CE), and Ibn Baṭṭūṭa (ms 1354 CE).14
For example, in the Middle Senegal Valley, the earliest historical evidence comes from the 1068 account by the Arab chronicler al-Bakri of the kingdom of Takrur, who mentioned two cities: Takrur and Silla. He also stated that people of both cities were Islamized by the king of Takrur. This made Takrur the first West African kingdom to accept Islam. Al Bakri reported that in the Middle Senegal Valley, the king of Takrou, War Jabi, converted to Islam early in the 11th century and waged war on non-Muslims.
This type of Islam, called militant Islam or expansionist Islam, was only found in the Middle Senegal Valley prior to the 18th century. Elsewhere, the pattern was a symbiosis between Islam and traditional religion, with only a small minority of committed Muslims. This type of Islam is called Islam de cours or accommodationist Islam.
Arab authors included the Middle Senegal Valley and the Inland Niger Delta in the same geographical area called Bilad al Sudan (the land of Black people). They stated that the Middle Senegal Valley and the Inland Niger Delta were involved in trade in staples, slaves, copper, and cotton. Moreover, the Senegal River and the Niger River were mistakenly called the Nil.
However, it must be noted that fewer details were provided on the Middle Senegal Valley towns and polities than on their counterparts in the Inland Niger Delta because Arab chroniclers were much more interested in the Empires of Ghana, Mali, Songhay, and Kanem- Bornu to the east. After the 14th century, Arab sources declined.
The second phase in Arabic sources is called the Ta’rikli phase (end of 16th century–17th century). It refers to an indigenous historical tradition in Arabic, written by African scholars linked to Islamic centers at Timbuktu and Jene. The two most famous examples are the Ta’rikli al-Fattach (by Mahmud Kati) and the Ta’rikn es-Sudan (by Abderrahman es-Sadi). These sources reflect the cultural vibrancy of these cities during that period. Now Africans have their voice, though unfortunately that voice has been too much influenced by the Islamic perspective. For instance during the Songhai empire, king Askia Mohamed was viewed more favorably by both Abderrahman es Sadi and Mahmud Kati than Soni Ali because he was considered as a pious and devoted Muslim who had good relationships with the Muslim clerics at Timbuktu.
The only non-African source during that second phase was Leon Africanus (1526) who described the culture and ways of life of West Africans. His major contribution remains his description of Timbuktu as a vibrant intellectual city, with numerous universities and scholars, where books are more valued than gold. His work transformed Timbuktu into a mythical city in the European imaginary and inspired the French explorer Rene Caille to aim at its discovery.
The Arabic phases of written history were followed by the French translation phase. During the colonial era oral histories and traditions became codified in writing thanks to the collaborative work between Maurice Delafosse (an ethnolinguist and colonial administrator), his father in law Octave Houdas (Oriental studies specialist), and Henri Gaden (ethnolinguist and colonial administrator).15 They translated chronicles from West African authors such as Sire Abbas Soh, Yoro Diaw, and Sheikh Moussa Kamara involving the history and culture of the people of Senegal.
However, oral and historical sources are clearly inadequate to give more than a general outline of major political shifts during the second millennium, and they offer no deeper insights into first millennium societies. Only archaeology can provide a perspective on the early development of societies.
There are two phases in the evolution of archeological research in the Middle Senegal Valley and the Inland Niger Delta: the colonial and postindependence phase, and the internationalization phase.
The colonial and postindependence phase: Anthropology in 19th-century Europe was dominated by social Darwinism and its corollaries: primitive theories, racial classification, and categorization. Overall, African reality, as it appeared in the literature during that time, was a product of Europe and for Europe. Africans were described in terms of helping Europeans understand themselves (the civilized) and their mission (the White Man’s Burden). Africa was constructed to shed light on western existence (as shown in the debate about the superiority or inferiority of African culture, the moral explanation of the White Man’s Burden and its economic aspect, and theories of primitivity and the superiority of western civilization).
The cultural difference between Africans and the others was perceived as a vertical differentiation. For proponents of social Darwinism such as Herbert Spencer, by traveling across space, we can see mankind’s evolution through time. It was thought that in Africa, the past was similar to the present. Africa was perceived as ahistorical. Africans were considered as representatives of a vanished past, indeed a living museum of the past.
That intellectual objectification of Africans was a corollary to their political subjection. Anthropology became used as a tool of domination. Africans became an object of anthropological knowledge to be used in a conspiracy theory to justify their domination. They were constructed in a way useful to the explanation and justification of the relations of power between Europe and Africa.
It was indeed believed that those who are different can be changed and integrated into the mainstream (e.g., the western) way of life. Furthermore, it was the obligation of Europe to do that. According to European statesmen such as Jules Ferry, Africans were uncivilized. It was the duty of Europe to civilize them.16 This came to be known as the White Man’s Burden and opened the door to the colonial conquest; hence, the future of Africa became controlled without the input of Africans (colonization).
That theoretical context affected the birth of archaeology in West Africa. Archaeology in the Middle Senegal Valley and in the Inland Niger Delta as well as in other areas of Africa started during the colonial period.17 As suggested by Holl, “African archaeology is the child of the colonial enterprise.”18 Early archaeological reports were influenced by the colonial project. Trigger summarizes it well:
African archaeology began in a colonial setting. It was first practiced by European visitors and settlers and the earliest interpretation of archaeological data reflected the understandings and prejudices that these people brought with them from Europe as they were modified to accord with economic and political conditions in which they found themselves. Many of these situations favored views of the past that saw Africans as incapable of self improvement and attributed any evidence of advanced cultures to the actions of civilized colonists who had come from elsewhere (1990).19
The earliest archaeological researches were undertaken by colonial administrators or army officers. Archaeology was under the supervision of the Institut Français d’Afrique Noire (IFAN). That research institute undertook research in all territories of the Afrique Occidentale Française (French-dominated West African countries) such as the Middle Senegal Valley and the Inland Niger Delta. However, the methods used were often not made explicit. Most of the artifacts came from surface collection or gullies or were found by chance during construction work; this random nature of the artifacts, added to the lack of description of the stratigraphy and the depositional sequence, cast doubt on any information or knowledge developed from their findings.
Theoretically, research on early West African cities was dominated by the “historical” school of archaeology. This approach was notably particularistic in scope and descriptive in intent. Archaeological questions of major interest have usually been framed in terms of the descriptions of West African towns and entrepots provided by Arab chroniclers and travelers from the 9th century AD onward. The identification of a particular archaeological site with a local place one mentioned in one or more sources was often a primary research objective.
Investigation therefore tended to focus on those aspects of site layout and architecture which correspond to details provided in the documents. Other classes of archaeological data collected likewise reflect an interest in validating and supplementing the historical sources; imports and trade goods were carefully retrieved and minutely described, to be marshaled as evidence for participation in the trade networks outlined by the early Arab chroniclers. Characteristically, fewer samples were obtained from lower levels corresponding to the preurban settlement. The Inland Niger Delta provided a good example. According to Togola, “No archaeological report on the Inland Delta prior to the late 1970 provided qualitative and quantitative data. This, added to the lack of stratigraphic information, precluded assessment of important issues such as change through time.”20
In fact, theoretically, until the 1970’s, diffusionist scenarios within a broad evolutionary tableau of simple to civilized dominated views of Africa’s past. Vaufrey’s notion of Africa as a “backward continent” had roots deep in the 19th century and was widely shared into the 1960s.21 Under the assumption of developmental stasis within Africa, the locus of change had to be sought elsewhere. Contacts with and diffusion from the historic civilizations of the Mediterranean world, Islamic cultures especially, were presumed to be the stimuli that led to social, economic, and political complexity in the Sudanic zone of West Africa. Hence researches were designed to find these artifacts linked to that stimulus: the fossil directeur. Consequently, this shackled African history within the Arabo-Berber stimulus paradigm. The best example for both the Middle Senegal Valley and the Inland Niger Delta is Delafosse (op. cit.) a French colonial administrator and ethnolinguist who attributed complexity and urbanization in these areas to Judeo-Syrians, Arabs, Berbers, and Carthaginians.
Mauny is another example. In his book Tableau Géographique de l’Ouest African au Moyen Age, he attempted a broad synthesis of West African history by combining written records (mainly the tarikhs), oral tradition, and archaeological data.22 His conclusion was that there was a lack of indigenous change and an implied need for European influence; like many before him, Mauny’s theory also froze the African past and judged African society to be static until the arrival of the Arabo-Berber, or Islamic civilization.23 Mauny considered the pre-Arabo-Berber contact period as an era characterized by a lack of innovation and progress. His theory did not help in understanding the indigenous development that led to the formation of the political entities that welcomed Arab and Berber traders. He did not shed light on the cultural and social dynamics, nor did he elaborate on the subsistence strategy prior to the trans-Saharan trade.
The internationalization phase: A major methodological shift occurs with the internationalization of research during the late 1970s in the Inland Niger Delta and the early 1990s in the Middle Senegal Valley. In the Inland Niger Delta, a Dutch team led by Bedeaux widens the research perspective by focusing on subsistence activities and by providing detailed stratigraphic information. They found the presence of domesticated African rice [oryza glaberrima], sorghum, pennisetum, and fonio, as well as domesticated animals such as cattle, goat, and sheep.24
Moreover, major works were undertaken by American archeologists, including Susan McIntosh and Roderick McIntosh at Jenne Jeno, Hambarketelo, and Dia. Their works were very important in terms of theoretical implications, research, and methodology. They used a processual approach and made it a cornerstone of their research strategy, a regional approach that involved controlled excavation, holistic analysis of the artifacts, and ceramic chronology established by several radiocarbon dates. The survey result shows that the Jené-jeno urban sphere was within a 4 km radius, enclosing 70 satellite sites, ten of which were excavated. Data collected shed new light on patterns of indigenous urbanism, trade networks, complexity, settlement organization, specialization, and subsistence strategies in the Inland Niger Delta.
They also conducted research at Timbuktu in 1984 also taken up during the 1990s by Timothy Insoll.25 Their research shows the absence of a Late Stone Age site, but also shows a clustering of villages starting 500 AD that was abandoned at 1500 AD. The importance of early trade was noticed: “. . . there are sound geomorphological reasons for believing that a major trade center could have existed at or near Timbuktu in the late first/early second millennium AD.”26
Susan McIntosh and Roderick McIntosh also did important work in the Middle Senegal Valley at the beginning of the 1990s. Their research leaned on some previous important work. In 1974, Martin and Becker pioneered a more systematic approach to survey, using a method that followed roads. Their research in the valley was part of a wider project that consisted of finding, mapping and classifying all archaeological sites found in Senegal and in the Gambia.27 They found more than three hundred archaeological sites that were published in their Repertoire des sites protohistoriques de la Sénégambie.28 They divided the archaeological sites found in Senegal and in Gambia into four types. Each type was linked to a specific geographical region: the megaliths (southern Senegal and the Gambia), the tumulus (Central Senegal), the shell mounds (Senegalese coast), and the tell sites (Middle Senegal Valley).
From that conceptual link between site typology and geographical location, they considered the Middle Senegal Valley to be a tell sites zone. That zone, called “zone de sites d’anciens villages,” means literally that they considered the type site found in the Middle Senegal Valley to be the remnants of ancient villages. Also, their division of Senegambia into four archaeological zones tended to associate each archeological area with an exclusive cultural mechanism (or coherent ethnic group). This does not, however, diminish the merit of their monumental work, which is still a sourcebook that helps fill some critical knowledge gaps. Their work is still useful for the Middle Senegal Valley archaeology as a point of departure.
Meanwhile more research projects emerged: Thiam and Marlhens-Niang focused on ethnoarchaeological analyses of the Middle Senegal pottery.29 Others looked at the surface remains of several iron smelting sites in both Senegal and Mauritania (where 60,000 furnaces were counted along a 15 km stretch of the river).30 Bocoum excavated at Tulel Fobo yielded information indicating an occupation during the second half of the first millennium.31
This period is marked by the addition of new methodological and theoretical approaches during an important project called the Middle Senegal Valley Project. The project, led by Susan McIntosh, Roderick McIntosh, and Hamady Bocoum provided a framework for approaching progressive issues, such as change through time in material culture, technology, and the development of specialization. The project methodology was based on excavation of deeply stratified sites (at Cubalel, Siwre, and Sincu Bara) and extensive mapping and survey. It was aimed at assessing the full range of sites and at providing data on the chronology of change in material culture, technology, subsistence, and trade.
Data from fourteen excavated sites yielded evidence for a ceramic sequence of four phases that was calibrated by over thirty-five radiocarbon dates. The C-14 series was also used to calibrate an archaeo-magnetic curve for the region.32 An archaeological survey provided data on settlement patterns, primarily on noninundated areas (100% coverage of levees) within the walo, although the jejeengol (transition zone between the floodplain and the upper lands) was also investigated. The survey team discovered 144 sites beyond the 36 sites already recognized by Martin and Becker.33 Of these, 79 were small, flat, ephemeral surface scatters known as plages, which were mapped but not surface collected.
Occupation Dynamics in the Middle Senegal Valley and in the Inland Niger Delta
The beginning of the occupation in the Inland Niger Delta and in the Middle Senegal Valley is linked to paleoclimate change. During the period from 4500 BP (2550 BC) through the first millennium the climate was characterized by high fluctuations. By 2500 BP (500 BC) the climate became arid. People left the Sahara following paleochannel corridors to move south (toward the Sahel) into the wetlands of the Middle Senegal valley and the Inland Niger Delta34 (Figure 5).
Occupation Dynamics in the Middle Senegal Valley
The earliest occupation evidence came from Walalde. Agropastoralists with a few sheep and goats first occupied Walalde sometime between 800 and 550 cal BCE.35 Because of the imprecision of the calibration curve, it is impossible to state whether their arrival was in the earlier or later part of this range. Occupation is initially sparse and episodic, with low artifact densities in the earliest levels, but a high degree of comminution of potsherds suggests exposure of material on the surface to trampling during periods of slow accumulation. The funerary practice consisted of wrapping the body in an organic shroud (grass mat or perhaps animal skin) that was set alight and burned very quickly and at relatively low temperatures. The bones were superficially charred in places where very little muscle separated skin from bone. Red ochre was scattered over the body.
The material culture of these early inhabitants consisted notably of iron artifacts and stone (hematite, laterite chunks, and fragmented pebbles) and bone, in addition to pottery and a variety of fired zoomorphic figurines. Most materials were locally available from within a 5–10 km radius. Only stone beads (cornaline and basalt) and chert were obtained over substantial distances. These exotics were quite rare. The early assemblage is provisionally assessed as transitional Late Stone Age. Two small pieces of slag are the only possible evidence for smelting in this early phase. Although the Walalde Phase I iron artifacts are among the earliest well-dated in situ iron in West Africa, there is no evidence to inform us about the context of their production or procurement.
The affinities of this early assemblage are suggested by the pottery, which shares specific similarities in vessel rim form and decoration to the Bouhdida assemblage (2600–2500 BP) in the region of Nouakchott. Bouhdida is presumed to be largely contemporaneous with the Akjoujt.
Intensity of site usage fell at the beginning of Phase II (550–200 cal BC). Accumulation was slow. Artifact and bones densities were fluctuating and low, suggesting episodic occupation. The inhumation of a male wearing an iron bracelet and copper earrings was immediately underlain by Level 20 and C-14 dated to c. 2500 BP. The source of the copper appears to be Akjoujt in Mauritania. The Phase II pottery assemblage is very similar to Phase I, suggesting that a single, evolving ceramic tradition is present. In the early part of Phase II (550–400 BC), it appears that ironworking is part of seasonal and episodic occupation.
Soon after the abandonment of Walalde, the occupation of Grand Kaskas began (150 cal BC–100 cal AD). The assemblages recovered from the Grand Kaskas sites likely represent a further continuation of the pottery tradition at Walalde. Herding of cattle, sheep, and goats continues to be important, but fishing represents a significant activity, evidenced by a large number of net weights (particularly at K2) and fish bones.
Almost four kilograms of slag was found at K1, plus eight tuyere fragments, suggesting considerable metallurgical activity. It is of particular interest that the C-14 dates of the deposits at three of the nine areas of archaeological deposits at Grand Kaskas were largely contemporaneous, suggesting that the former large mound complex consisted of a number of simultaneously, rather than sequentially, accumulating mound deposits. We can now state that populations aggregated (seasonally?) on a 5+ ha cluster of sites from at least the turn of our era.
This development corresponds closely with the environmental crisis at 1900 BP, when severe drought reduced the flow of the river so dramatically that salt water flowed upriver for 300 km from its mouth.36 Kaskas may represent the spatial integration of specialist economies (herding, fishing) to achieve greater productivity in a situation of extreme environmental stress.
The most reliable information concerning the two millennia AD came from the MSV project. Data from fourteen excavated sites at Cubalel and Siwre yielded evidence for a four-phase ceramic sequence that was calibrated by over thirty-five radiocarbon dates that were also used to calibrate an archaeo-magnetic curve for the region.37
Copper and other exotics were found only at sites with post-900 AD assemblages. Furthermore, there was no sign of an emergent site hierarchy. Nearly all pre-900 AD sites were 2 ha or less in size,38 a pattern that contrasts with the large-scale settlements that grew throughout the first millennium in the Malian Inland Delta.
One of the main conclusions of that multistage research program was that: “The preliminary results of the survey confirm the general picture provided by excavation: small scale societies that underwent little change throughout the first millennium. The evidence suggests that MSV societies underwent a major and extremely rapid transformation in scale and complexity between the 10th century and the period of the Takrur Empire, which is historically attested just over 100 years later.”39
Occupation Dynamics in the Inland Niger Delta
The occupation stared around the end of the last millennium BC. During phase I/II (250 BC–400 AD), the site was what its cluster region had been measured as already, an area of 25 ha. Sites were different in size and in function. The pottery was fine and sand-tempered. No Late Stone Age evidence was found. However, early occupants practiced metallurgical activities. Their diet was based on fish and cattle. Bones of catfish and Nile perch were found in the early deposits. They also ate a variety of local wild grasses, domestic cow, dwarf goat, and wild bovids.40 Where do the early occupants of the Inland Niger Delta come from? It is worth noting that the pottery shows affinities with the Sahara.
During Phase II, people cultivated rice, sorghum, and millet. Domesticated rice was found only from Phase II onwards. This leaves still open the process of rice domestication in West Africa. Clay figurines were also associated with Phase II levels. According to Susan and Roderick McIntosh they resemble those found at Karkarichinkat and Daima made by Late Stone Age pastoralists/collectors.41
During Phase III (AD 400–AD 850), Jenne Jeno and neighboring Hambarketelo reached their maximum extension (41 ha). Jenne Jeno was surrounded by twelve other occupied sites. A massive city wall had been built. Evidence of imported and status commodities are found: gold, cooper. The trend towards site increase was found in other areas of the Inland Niger Delta. At Timbuktu, Middle sites (AD 500 to AD 1500) reached their maximum extension. The subsistence is still based on rice, fish, and bovid.
The last phase is Phase IV (850–1400 AD). It is associated with an architectural shift from circular to cylindrical mud brick and with the presence of a defensive wall ten feet wide and almost one mile long. Phase IV corresponds to the decline and abandonment of the site and coincides with North African artifacts and influences such as spindle whorls, glass beads, and rectangular mud building plans.42 The subsistence remains based on rice, fish, and bovid.
As summarized by Susan McIntosh:
Terra cotta statuettes appear in domestic contexts embedded in house foundations. Bronze is present initially, replaced by brass from north Africa c. 1000. This coincides with appearance of North African glass beads, spindle whorls, and rectilinear houses in cylindrical mud brick. Statuettes become more elaborate and stylistically varied. Among these, warrior and mounted warrior styles emerge. Burial continues in urns and by simple inhumation. Levels of dental pathology appear to increase. Grave goods are absent. From c. AD 1200, settlement retracts until the site is completely abandoned by AD 1400 (1999).43
The abandonment was due to climatic or cultural factors. It started around 1100 AD at Hambarketolo and Kaniana and ended with the abandonment of Jenne Jeno at 1400 AD. The pattern of depopulation is found in other areas such as Timbuktu where sites became smaller from 1500 AD through 1900 AD.
The material culture and the clustering aspect of its regional occupation was also found at Dia (100 km northwest of Jenne) showing an economic and cultural integration in a wider region.
At the end of Phase IV, a new ecological zone (the sandy upland that favored by millet and sorghum cultivators) became the preferred one. This driving force behind this process was Bambara millet farmers.
Both the Middle Senegal Valley and the Inland Niger Delta were occupied during a regional process of abandonment of Saharan settlements in a context of drought and paleoclimate deterioration during the end of the second millennium and beginning of the first millennium BC. Potteries of both areas show affinities with the Sahara. However, the occupation process in the Middle Senegal Valley seems to have started a few centuries earlier than in the Middle Niger.
After initial occupation, the two areas took different trajectories in specialization and settlement dynamics. The following comparative analysis between the floodplains was done by Susan McIntosh (Table 1).44
Table 1. Comparison of the inland Niger Delta at Jenné-jeno and the Middle Senegal Valley at Cubalel c. AD 800 (in S. McIntosh 1999, p. 19).
Inland Niger delta
Middle Senegal Valley
clustered hierarchy cluster
<1000 (does not include seasonal population on floodplain)
over 600 km
no grave goods
no grave goods
In the Middle Senegal valley copper and other exotics were found only at sites with post-900 assemblages. Furthermore, there was no sign of an emergent site hierarchy. Nearly all pre-900 AD sites were 2 ha or less in size, a pattern that contrasts with the large-scale settlements that grew throughout the first millennium in the Malian Inland Delta where the first millennium is associated with trade, social stratification, and site hierarchy and growth. Besides showing evidence of indigenous urbanism and trade that preceded the transSaharan trade, “the homogeneity of material culture throughout a large area of the Inland Delta from 400–1400 AD may indicate a degree of administrative control, ensuring a stable political environment for exchange.”45
By the 9th century both areas are in contact with Islam and the trans-Saharan trade. The form of Islam in the Middle Senegal was militant while that in the Inland Niger Delta was less political (Islam de cours). By the second millennium, the Middle Senegal valley underwent a major change in scale and complexity.
Discussion of the Literature
In comparative arid-lands floodplain prehistory, from both the data-collection (culture-historical) and the theoretical aspects, Africa is a good example of an area left out of the archaeological theorizing on complexity. Susan McIntosh explains the peripheral position of the African data in the standard complexity canon by the fact that until the 1970s, research on complexity done in West Africa used a diffusionist paradigm that considered the Mediterranean as the stimulus center for African change, technology, and innovation.46 Fortunately, since the 1970s, the neoevolutionary approach that often pictures Africa as a backward continent has been challenged. Africa has played a prominent role in that challenge with the finding of large complex settlements.
The finding of new forms of complexity in Africa that were not included in the range of expressions predicted by the standard complexity theories should be another reason for the inclusion of African data in any archaeological theorizing on complexity. On the basis of ethnographic and historical data, we note the existence in Africa of societies with a central authority balanced within a diffuse, segmented and heterarchical power structure. There are also societies with horizontal differentiation and consensus-based decision making.
All these types of organization are based on the cohabitation between several sources of power vested in corporate entities, such as lineages, age groups, cults and secret societies. These new findings have set the stage for Susan McIntosh to propose new pathways to complexity. She states that Africa “provides a rich corpus of material relevant to an understanding of societies in which central authority, often a ritual nature, is paired with a powerful structure that is diffuse, segmentary, and heterarchical, as well as societies in which considerable complexity is achieved through horizontal differentiation and consensus-based decision making. The distribution of power among several corporate entities (e.g., lineages, secret societies, cults, age grades) can be regarded as a strategy that has successfully resisted in a variety of ways the consolidation of power by individuals.”47
She makes the case that we must take into consideration horizontal complexity; that is, the existence of several power sources within the same polity. Her proposition joins that of proponents of heterarchy theory who call for serious investigation of situations where several competing sources of power hold more-or-less equal sway in the political arena.48 Africa is a fertile ground for that form of theoretical endeavor.
Encouraging results have started to come from Africa, particularly from West Africa. The Middle Niger and the Middle Senegal Valley can offer insights into various trajectories toward specialization, population heterogeneity, and other new forms of complexity. These two floodplains have demonstrated that the Middle Senegal Valley is a prime candidate for illuminating issues of broad comparative interest. Among these issues are the circumstances and the forms of the transition from the Late Stone Age to the Iron Age, technological innovation and elements of landscape used by subsistence groups and by artisan producers, and emergence of social complexity. Indeed, in the Middle Senegal Valley, as well as in the Inland Niger delta, occupation seems to have started soon after the abandonment of Saharan Late Stone Age settlements. Hence the Middle Senegal Valley might shed light on the processes that led to transition from Late Stone Age to Iron Age, the emergence of iron-using societies, and the emergence and development of iron metallurgy itself.
These are also very important works on internal processes of change as well as on power, gender, knowledge, and population dynamics, important questions that are also being illuminated using additional data from bioanthropology (genofacts), isotopic analysis, and a better calibration curve in order to have a more precise chronology.
There are two types of sources: oral and written (Arabic).
Oral sources are endogenous. Considered as the living voice of past Africa, the oral tradition reflects what Africans remember about their past and how that past is used through the symbolic reservoir to shape the present and to influence the construction of the reality. However these oral sources are sometimes very limited in scope because quite often they reflect the point of the local elite. Moreover, the oral tradition was codified during the colonial period by educated Africans trained in French schools. Quite often their history was meant to fulfill the French historiographical colonial order (see Yoro Diaw and Sire Abas Soh). Their writing was meant to fit the colonial project and one its corollaries: the primitive theories and the diffusionist paradigm. The origin of West African states was thought to be from Ancient Egypt or from the Middle East. Urbanization and complexity in West Africa were explained as being cause by contact with the Berbers, and the Arabo-Islamic world.
Written sources include accounts by geographers, travelers, and scholars from North Africa, Spain, and the Middle East. Written in Arabic, they are considered as “the most important written documents on the history of the Western Sudan for the period that extends roughly from the middle of the eight century AD to the middle of the second millennium AD.”49 The best known authors were, among others, al Masudi al Bakri, Al Idrissi, al Umari, Ibn Battouta, Hassan Ibn Mohammad al-Wuzza’n (best known as Leo Africanus), and Ibn Khaldum.50 However, these sources are sometimes confusing when it comes to the name of people and place. They contain exaggerated facts, show sometimes a lack of understanding of local culture, and occasionally utilize secondary sources without critical thinking. Moreover, the sources were also often influenced by the Arabic and Islamic background of their authors; certain descriptions of African culture and African traditional leaders were Arabo-Islamic-centric.
There were also African Muslims who used Arabic scripts to codify oral history. Called tarikhs (chronicles), these histories were written by local Muslim scholars from famous Islamic centers such as Timbuktu, Jenne, and Dia. The two most famous tarikhs are: Ta’rīkh al- Fetach, written by three generations of the Kati family between 1519 and 1665, and Ta’rīkh al- Sūdān, written by es Sadi between 1627 and 1665. Their approach was descriptive with a narrative writing style.
Bocoum, Hamady. La métallurgie du fer au Sénégal: approche archéologique, technologique et historique. Thèse 3eme cycle. Université Paris I, 1986.Find this resource:
Boutillier J. L. La Moyenne Vallée du Sénégal; étude socio-économique. Ministère de la Coopération (République française), I.N.S.E.E., Service de Coopération. Paris: Presses Universitaires de France, 1962.
Brooks, George. “A Provisional Historical Schema for Western Africa based on Seven Climatic Periods (c. 9,000 BC to the 19th century).” Cahiers d’Etudes Africaines, 26 (1986): 43–62.Find this resource:
Brunet-Moret, Yves, Pierre Chaperon, Jean Pierre Lamagat et al. Vols. 1 and 2 of Monographie du Niger. ORSTOM, Collection Monographies Hydrologiques Numéro 8. Paris: Editions ORSTOM, 1986.Find this resource:
Dème, Alioune. Evolution Climatique et processus de mise en place du peuplement dans l’Ile A Morphil. Mémoire de Maitrise, Université Cheikh Anta Diop, 1991.Find this resource:
Dème, Alioune. “Archaeological Investigations of Settlement and Emerging Complexity in the Middle Senegal Valley.” PhD dissertation, Rice University, 2003.Find this resource:
Dème, Alioune, and Susan K. McIntosh. “Excavations at Walalde: New Light on the Settlement of the Middle Senegal Valley by Iron-Using Peoples.” Journal of African Archaeology 4, no. 2 (2006): 317–347.Find this resource:
Diop, Cheikh. The African Origin of Civilization: Myth or Reality. New York: Lawrence Hill, 1974.Find this resource:
Gallais, Jean. Le delta intérieur du Niger, étude de géographie régionale. Vol. 1 and 2. Dakar, Senegal: IFAN, 1967.Find this resource:
Gallais, Jean. Hommes du Sahel. Coll. Géographes. Paris: Flammarion, 1984.Find this resource:
Goudiaby, Assane. “L’évolution de la pluviométrie en Sénégambie de l’origine des stations à 1983.” Mémoire de Maitrise, Université Cheikh Anta Diop, 1984.Find this resource:
Holl, Augustin. West African archaeology: Colonialism and Nationalism. In A History of African Archaeology edited by Peter Robertshaw, 196–308. London: J. Currey, 1990.Find this resource:
Leroux, Marcel. “Le climat de l’Afrique Tropicale.” Thèse d’Etat, Université de Dijon, 1980.Find this resource:
Lézine, Anne-Marie. “Late Quaternary Vegetation and cClimate of the Sahel.” Quaternary research 32 (1989): 317–334.Find this resource:
Lézine, Anne-Marie, and Joel Casanova. “Pollen and Hydrological Evidence for the Interpretation of Past Climates in Tropical West Africa during the Holocene.” Quaternary Research 8 (1989): 45–55.Find this resource:
Marlhens-Niang, Gabrielle. “La céramique du site ancien d’Akhréjit (RIM): son évolution et son extension au Sénégal.” Thèse de doctorat 3eme cycle, Paris I, Sorbonne 1986.Find this resource:
Martin, Victor, and Charles Becker. Répertoire des sites protohistoriques du Sénégal et de la Gambie. Kaolack, Senegal: Ronéotypé, 1974.Find this resource:
Martin, Victor and Charles Becker. Inventaire des sites protohistoriques de la Sénégambie. Kaolack, Senegal: CNRS, 1984.Find this resource:
Mauny, Raymond. Tableau géographique de l’ouest africain au Moyen Age d’après les sources écrites, la tradition et l’archéologie. Mémoire de l’Institute français Afrique Noire 61. Dakar, Senegal: IFAN: 1961.Find this resource:
McIntosh, Roderick. “Floodplain Geomorphology and Human Occupation of the Upper Inland Delta of the Niger.” The Geographical Journal 149, no. 2 (1983): 182–201.Find this resource:
McIntosh, Roderick. “The Pulse Model: Genesis and Accommodation of Specialization in the Middle Niger”. Journal of African History, 34, no. 2 (1993): 181–201.Find this resource:
McIntosh, Roderick. “Before Timbuktu: Cities of the Elder World.” In The Meanings of Timbuktu. Edited by Shamil Jeppie and Souleymane Bachir Diagne, 31–44. Dakar, Senegal: CODESRIA/HSRC, 2008.Find this resource:
McIntosh, Roderick J., and Susan McIntosh. “Archaeological Reconnaissance in the Region of Timbuktu, Mali.” Final report to National Geographic Research 2 & 3, 1985.Find this resource:
McIntosh, Roderick, and Susan McIntosh. “From siècles obscurs to Revolutionary Centuries on the Middle Niger.” World Archaeology 20, no. 1 (1988): 141–165.Find this resource:
McIntosh, Roderick, Joseph Tainter, and Susan McIntosh (eds). The Way the Wind Blows: Climate, History and Human Action. New York: Columbia University Press, 2000.Find this resource:
McIntosh, Susan. Changing perceptions of West Africa’s Past: Archaeological research since 1988. Journal of Archaeological Research 2 (1994): 165–198.Find this resource:
McIntosh, Susan. “A tale of two floodplains: comparative perspectives on the emergence of complex societies and urbanism in the Middle Niger and Senegal Valleys.” Paper presented at the Proceedings of the Second World Archaeological Congress Intercongress, Mombasa, 1999.Find this resource:
McIntosh, Susan, and Roderick McIntosh. “Initial Perspectives on Prehistoric Subsistence in the Inland Niger Delta (Mali).” World Archaeology 11, no. 2 (1979): 227–243.Find this resource:
McIntosh, Susan. and Roderick McIntosh. “Current Directions in West African Prehistory.” Annual Review of Anthropology 12 (1983): 215–258.Find this resource:
McIntosh, Susan, and Roderick McIntosh. “The Early City in West Africa: Towards an Understanding.” The African Archaeological Review 2 (1984): 73–98.Find this resource:
McIntosh, Susan, Roderick McIntosh, and Hamady Bocoum. “The Middle Senegal Valley Project: Preliminary Results from the 1990–1991 Field Season.” Nyame Akuma 38 (1992): 47–61.Find this resource:
Michel, Pierre. Les bassins des fleuves Sénégal et Gambie: Étude géomorphologique. Mémoire ORSTOM 63. Paris: ORSTOM, 1973.Find this resource:
Monteillet, Jean, Henri Faure, and Paolo Antonio Pirazolli, et al. “L’invasion saline du Ferlo (Sénégal) à l’Holocène Supérieur (1900 BP).” Paleoclimatology of Africa and the Surrounding Islands 13 (1981): 205–215.Find this resource:
Niang, Madické. “La vallée alluviale du Sénégal: Le milieu bioclimatique.” Thèse de 3eme cycle. Université Cheikh Anta Diop, 1973.Find this resource:
Nicholson, Sharon E. “Rainfall and Atmospheric Circulation during Drought Periods and Wetter Years in West Africa.” Monthly Weather Review 109, no. 10 (1981): 2191–2208.Find this resource:
Nicholson, Sharon E. “Recent Rainfall Fluctuations in Africa and Their Relationship to Past Conditions over the Continent.” The Holocene 4, no. 2 (1994): 121–131.Find this resource:
Petit Maire, Nicole, eds Sahara ou Sahel ? Quaternaire récent du bassin de Taoudéni. Marseille, France: Imprimerie Lamy, 1983.Find this resource:
Robert-Chaleix, Denise, and Mamadou Sognane. “Une industrie métallurgique ancienne sur la rive mauritanienne du fleuve Sénégal.” Mémoire de la Societé des Africanistes 9 (1983): 45–62.Find this resource:
Thiam, Mandiomé. “La céramique au Sénégal: archéologie et histoire.” Thèse de doctorat 3eme cycle, Paris I, Sorbonne 1991.Find this resource:
Togola, Tereba. “Archaeological Investigation of Iron Age Sites in the Mema Region, Mali (West Africa).” PhD diss., Rice University, Houston, 1993.Find this resource:
Togola, Tereba. “Memories, Abstraction and Conceptualization of Ecological Crisis in West African Sahel.” Paper presented at the September climate workshop. Rice University, Houston, 1995.Find this resource:
Trigger, Bruce G. “The History of African Archaeology in World Perspective.” In A History of African Archaeology. Edited by Peter Robertshaw, 309–319. London: J. Currey, 1990.Find this resource:
Zwarts, Leo, and Mori Diallo. Eco-hydrologie du Delta. In Delta Intérieur du Niger: Ecologie et gestion durable des ressources naturelles. Edited by Eddy Wymenga, Bakary Kone, Jan van der Kamp, and Leo Zwarts, 45–63. Mali: Wetlands International, 2002.Find this resource:
(1.) Sharon E. Nicholson, “Recent rainfall fluctuations in Africa and their relationship to past conditions over the continent”, The Holocene 4. no. 2 (1994): 121–131.
(2.) George E. Brooks, A Provisional Historical Schema for Western Africa Based on Seven Climate Periods (c .9000 BC to 19th Century) (Madison, WI: African Studies Association, 1986); Nicholson, “Recent Rainfall Fluctuations”; Anne-Marie Lézine et Joel Casanova, “Pollen and Hydrological Evidence for the Interpretation of Past Climates in tropical West Africa during the Holocene”, Quaternary Science Reviews 8, no. 1 (1989): 45–55; N. Petit-Maire et al., éd., Sahara ou Sahel? Quaternaire récent du bassin de Taoudenni, Mali (Paris: Laboratoire de géologie du quaternaire du Centre national de la recherche scientifique, 1983).
(3.) Roderick J. McIntosh, Joseph A. Tainter, and Susan Keech McIntosh, éds., The Way the Wind Blows: Climate Change, History, and Human Action (New York: Columbia University Press, 2000).
(4.) P. Elouard, “Etudes géologiques et hydrogéologiques des formations sédimentaires du Guebla mauritanien et de la vallée du Sénégal.” (Paris: Imprimerie Louis-Jean, 1962).
(5.) Pierre Michel, “Les bassins des fleuves Sénégal et Gambie: Etude géomorphologique”, 1973.
(7.) Yves Brunet-Moret et al. Monographie hydrologique du fleuve Niger. Institut français de recherche scientifique pour le développement en coopération (Paris: OSTROM, 1986).
(9.) Susan Keech McIntosh and Roderick J. McIntosh, “Initial Perspectives on Prehistoric Subsistence in the Inland Niger Delta (Mali)”, World archaeology 11, no. 2 (1979): 197.
(10.) Roderick J. McIntosh, “Floodplain Geomorphology and Human Occupation of the Upper Inland Delta of the Niger”, The Geographical Journal 149, no. 2 (1983): 182–201. Retrieved from https://doi.org/10.2307/633602; and Susan Keech McIntosh et Roderick J. McIntosh, Prehistoric investigations in the region of Jenne, Mali: A study in the development of urbanism in the Sahel, vol. 89 (BAR, 1980).
(11.) Leo Zwarts et M. Diallo, 45–63.
(12.) Roderick J. McIntosh, “The Pulse Model: Genesis and Accommodation of Specialization in the Middle Niger”, The Journal of African History 34. no. 2 (juillet 1993): 181–220; Tereba Togola, “Memories, Abstractions, and Conceptualization of Ecological Crisis in the Mande World”, in The Way the Wind Blows: Climate, History, and Human Action, 181–192.
(13.) McIntosh and McIntosh, “Initial Perspectives on Prehistoric Subsistence in the Inland Niger Delta (Mali).”
(14.) John F. P. Hopkins et Nehemia Levtzion, Corpus of early Arabic sources for West African history (Markus Wiener Publishers, 2000); Joseph M. Cuoq et Raymond Mauny, “Recueil des sources arabes concernant l’Afrique occidentale du VIIIe au XVIe siècle”, 1975, Retrieved from http://archives.umc.edu.dz/handle/123456789/109224; and Raymond Mauny, Textes ct documents relatifs à l’histoire de 1’ Afrique: extraits tirés d’lbn Batuta, vol. 9 (section d’histoire de la Faculté des Lettres et Sciences Humaines, 1975).
(15.) Maurice Delafosse, Haut-Sénégal, Niger, vol. 2 (Paris: E. Larose, 1972); Henri Gaden, “Légende et Coutumes Sénégalaises: Cahiers de Yoro Dyao” (Paris: E Larose, 1912).
(16.) Jules Ferry before the French parliament: “It must be said openly that the superior races have a right with respect to the lower races. I repeat that there is a right for superior races, because there is a duty for them. They have the duty of civilizing the lower races” (July 28, 1885).
(17.) Augustin Holl, “West African Archaeology: Colonialism and Nationalism”, A History of African Archaeology, 1990, 296–308; McIntosh, “Floodplain Geomorphology and Human Occupation of the Upper Inland Delta of the Niger”; and Bruce G. Trigger, “The history of African Archaeology in World Perspective”, A History of African Archaeology 315 (1990).
(18.) Holl, 296.
(19.) Trigger, 310.
(20.) Togola, 19.
(21.) Vaufrey, Préhistoire de l’Afrique. vol. 2, au nord et à l’est de la grande forêt, 1969.
(22.) Raymond Mauny, Tableau géographique de l’Ouest africain au Moyen-Âge d’après les sources écrites, la tradition et l’archéologie (Dakar, Senegal: IFAN, 1961).
(23.) Mauny, 54.
(24.) Togola, “Memories, Abstractions, and Conceptualization of Ecological Crisis in the Mande World”, 20.
(25.) McIntosh, Roderick. “Before Timbuktu: Cities of the Elder World.” In The Meanings of Timbuktu, eds. Shamil Jeppie and Souleymane Bachir Diagne, 31–44 (Dakar, Senegal: CODESRIA/HSRC, 2008).
(27.) Martin, Victor, and Charles Becker, Répertoire des sites protohistoriques du Sénégal et de la Gambie (Kaolack, Senegal: Ronéotypé, 1974).
(28.) Victor Martin and Charles Becker, Inventaire des sites Protohistoriques de la Sénégambie (Paris: KAOLACK-C.N.R.S., 1984).
(29.) Thiam, Mandiomé, “La céramique au Sénégal: archéologie et histoire.” Thèse de doctorat 3eme cycle, Paris I, Sorbonne 1991; and Marlhens-Niang, Gabrielle. “La céramique du site ancien d’Akhréjit (RIM): son évolution et son extension au Sénégal.” Thèse de doctorat 3eme cycle, Paris I, Sorbonne 1986.
(30.) Denise Robert-Chaleix and Mamadou Sognane, “Une industrie métallurgique ancienne sur la rive mauritanienne du fleuve Sénégal,” Mémoire de la Societé des Africanistes 9 (1983): 45–62.
(31.) Hamady Bocoum, La métallurgie du fer au Sénégal: approche archéologique, technologique et historique. Thèse 3eme cycle. Université Paris I, 1986.
(32.) Susan Keech Mcintosh, Roderick J. McIntosh, et Hammadi Bocoum, “The Middle Senegal Valley Project: Preliminary Results from the 1990–91 Field Season”, Nyame Akuma 38 (1992): 47–61.
(33.) Victor Martin and Charles Becker, Inventaire des sites protohistoriques de la Sénégambie (Kaolack, Senegal: CNRS, 1984).
(34.) McIntosh, « The Pulse Model », 216.
(35.) Alioune Dème, “Archaeological Investigations of Settlement and Emerging Complexity in the Middle Senegal Valley” PhD diss., Rice University, 2003; and Alioune Dème et Susan Keech McIntosh, “Excavations at Walaldé: New Light on the Settlement of the Middle Senegal Valley by iron-Using Peoples”, Journal of African Archaeology 4, no. 2 (2006): 317–347.
(36.) J. Monteillet et al., “L’invasion saline du Ferlo (Sénégal) à l’Holocène supérieur (1900 BP)”, Paleoecology of Africa 13 (1981): 205–215.
(37.) McIntosh, McIntosh, et Bocoum, “The Middle Senegal Valley Project.”
(38.) Alioune Dème, “Evolution climatique et processus de mise en place de peuplement dans l’île à Morphil”, Unpublished Mémoire de maitrîse, Université de Dakar-Cheikh in Anta Diop, 1991.
(39.) McIntosh, McIntosh, and Bocoum, “The Middle Senegal Valley Project”, 57.
(40.) Susan Keech McIntosh et Roderick J. McIntosh, “From Stone to Metal: New Perspectives on the Later Prehistory of West Africa”, Journal of World Prehistory 2. no. 1 (1988): 145.
(41.) McIntosh et McIntosh, “Initial Perspectives on Prehistoric Subsistence in the Inland Niger Delta (Mali)”, 233.
(42.) Roderick J. McIntosh, “Archaeological Recon-naissance in the Region”, National, Geographic Research 2, no. 3 (1985): 302–319.
(43.) Susan Keech McIntosh, “A tale of two floodplains: comparative perspectives on the emergence of complex societies and urbanism in the middle Niger and Senegal valleys”, in Proceedings of the Second World Archaeological Congress Intercongress, Mombasa, P. Sinclair, ed. Published on the Uppsala University website http://www.%20arkeologi.%20uu.%20se/afr/projects/BOOK/Mcintosh/mcintosh.%20Htm, 1999, http://www.musik.uu.se/digitalAssets/9/9586_McIntoshAll.pdf.
(44.) Susan McIntosh, “A tale of two floodplains: comparative perspectives on the emergence of complex societies and urbanism in the Middle Niger and Senegal Valleys.” Paper presented at the Proceedings of the Second World Archaeological Congress Intercongress, Mombasa, 1999.
(45.) Dème, “Evolution climatique et processus de mise en place de peuplement dans l’île à Morphil”; McIntosh, McIntosh, et Bocoum, “The Middle Senegal Valley Project.” McIntosh, “Floodplain Geomorphology and Human Occupation of the Upper Inland Delta of the Niger”, 246.
(48.) Carole L. Crumley, “A dialectical critique of hierarchy”, Power relations and state formation, 1987, 155–169; and Carole L. Crumley, “Heterarchy and the analysis of complex societies”, Archeological Papers of the American Anthropological Association 6, no. 1 (1995): 1–5.
(49.) Tereba Togola, “Archaeological Investigations of Iron Age sites in the Mema region, Mali (West Africa)” PhD diss., Rice University, 1993, 5.