Early polities of the Western Sudan

DAVID C. CONRAD

From the late eighth century to 1500, a series of major polities emerged in the Sudanic zone between the Niger River and the Atlantic. All drew significant resources from long-distance and trans-Saharan trade in gold, salt and slaves, according to contemporary chroniclers writing in Arabic.

The history of Western Sudanic states during our chronological period centers on three main polities. The earliest was the Soninke kingdom of Wagadu/ Ghana (not to be confused with the modern coastal state of that name) that emerged during the period 500-700, and was in decline by the second half of the eleventh century. Following a period of political turmoil of unknown duration, most likely in the first half of the thirteenth century, Manding chiefdoms in the region of the Upper Niger and its tributaries unified into what would become the Mali Empire that attained its apogee in the fourteenth century and was in decline by the mid-fifteenth century. By c. 750-950, the kingdom of Gao farther east on the Middle Niger Bend had become sufficiently prosperous as a terminus for trans-Saharan trade that it expanded into the Songhay Empire by the 1490s. Songhay replaced Mali as the dominant power in the Western Sudan until 1591, when it was conquered by an invading Moroccan army.

This chapter is dedicated to the memory of Nehemia Levtzion. in more detail, but will first discuss the role of oral traditions and recent interpretations of archaeological findings in West African history.

The importance of oral tradition for West African history

West African oral tradition rarely reveals facts about the past that can be accepted according to strict European historiographical standards, because the existence of people and events that are described in the narratives usually cannot be confirmed by independent evidence such as written documenta­tion involving contemporary eyewitnesses. Nevertheless, oral tradition does convey useful information, both literally and metaphorically, about African perceptions of the past and how those perceptions affect present-day cultural values. In many rural West African communities, traditional social values and customs have not been submerged by foreign influence, as has been the case in capital cities and other urban centers. In agricultural villages and market towns of the countryside, notions of what happened in the distant past are still expressed through oral traditions that provide part of the framework for the way people live from day to day. The deeds of ancestral heroines and heroes are described in narratives that are passed down from one generation to the next in chains of oral communication that can be measured in spans of time ranging from several generations to many centur­ies. In some cases, the discourse rises to the level of epic that is filled with rich and colorful imagery in both narrative and song, reflecting the values of the cultures that produced them. Epic narratives vary in length, content and complexity depending on the desires and composition of the audience, and on the knowledge, purpose, and momentary whim of the performer. They tell the stories of village communities, kingdoms and empires that were peopled by charismatic leaders both male and female, who founded family lineages and accomplished momentous deeds during periods that are recalled as defining moments of the past.

Populations descended from the medieval West African states boast unsurpassed levels of oral art and consciousness. In the epic narrative of the Manding and some culturally related societies, the ancestors remain as they have apparently been for many centuries, the means of identification for every stranger entering a town or village, and for every householder receiv­ing the strangers. The ancestors are evoked during any serious greeting, they are regularly praised in songs by the ubiquitous bards, their spirits are present at every council meeting, and sacrificial offerings are made to them before any serious project is undertaken. It is no exaggeration to say that the ancestors described in kuma koro or “ancient speech” define the identity of each and every member of Mande society, regardless of gender. Therefore, if the ancestors were to be forgotten, the fundamental essence of indigenous values - as perceived by the people themselves - would disappear. For foreign scholars to maintain that it is not possible for, say, the hero Sunjata and his companions to be known after so many centuries and that they and their deeds are entirely mythical (that theirs is “imaginary history” as one writer has phrased it), is to deny a fundamental fact of an extraordinarily vibrant culture: genuine reverence for the ancestors has kept certain aspects of the past more alive than could have been accomplished with the written word.

Cities and trade as a basis of state power

In the late 1970s, some important strides forward in the acquisition of archaeological data in West Africa were made when an archaeological team first visited the site ofJenne-jeno in the Middle Niger Delta of the republic of Mali. They were impressed by the vastness of an ancient settlement littered with ruins of clay-brick houses, exposed and still occupied burial urns, evidence of extensive iron works, and many tons of broken pottery among other surface artifacts. However, in what appeared to be an ancient urban landscape (later dated to c.

Nowadays it is evident that monumental manifestations of state power are absent in ancient settlements of the Middle Niger because they evolved without the kinds of power hierarchies and accompanying symbols charac­teristic of urban centers of the ancient Middle East and elsewhere. One developing argument is that the more familiar kinds of despotic, state-driven societies lack sustainability, being fragile and prone to eventual collapse because of their rigid hierarchical infrastructure. In the Middle Niger Delta, archaeologists found a different species of ancient urbanism, with non- despotic communities of diverse corporate groups in no need of a king or other indications of a state-based city. Influenced by the unpredictable and potentially hazardous sub-Saharan Middle Niger environment, these commu­nities spontaneously evolved into what Roderick McIntosh calls a “self­organizing landscape.” Passing through multiple stages of development, the process culminated by the middle of the first millennium ce in a kind of “multi-satellite, multi-corporate urban complex” that was innovative and flexible enough to achieve the type of long-term sustainability that the McIntosh expeditions found at Jenne-jeno and other partly explored locations in the Middle Niger Delta.¹

Research of the last several decades indicates that sociopolitical develop­ment during the first millennium ce was influenced by a wide range of economies involving various degrees of mobility.

The antiquity of both trans-Saharan trade and the trade-based polities of the Western Sudan has been the subject of considerable discussion. Some scholars argue for a trade route from North Africa to the Niger Bend in the mid-first millennium bce. Recoveries of Roman coins are limited to only three sites in southern Mauritania, but Kevin MacDonald has presented evidence on behalf of “a pre-existing network of down-the-line exchange” in semi-precious stones.^{[668] [669]} Susan McIntosh has outlined a significant amount of evidence for a variety of regional exchange systems throughout the southern Sahara and Sahel. According to her archaeological findings at Jenne-jeno, the earliest unambiguous evidence discovered to date for the re-establishment of links between desert and savannah is the presence of copper by the fifth century ce, which also indicates the likelihood of a desert salt trade.^[670] Discussion of trade routes has of course included the question of when gold began to be transported along the Saharan networks to reach North Africa.

Imports of copper and glass in sites just south of the Sahara increased during the period 400-800. The latter half of this period coincides with the initial Arab advance into North Africa, and the establishment of trading centers at Tahert and Sijilmasa in the Maghreb. Large-scale transport of heavy commodities such as salt and metals was made possible by the spread of the camel as a desert pack animal in the first millennium ce. With the spread of Islam and literacy, more expansive trade networks carrying far more goods were in place by 1000. In the eleventh century, a desert confeder­ation of the desert-dwelling Berbers known as Sanhaja established the powerful Almoravid Empire, which provided a unified political field for trade from the Senegal River Valley to the Maghrib (i.e. northwestern Africa), southern Spain, and Egypt.

Wagadu/Ghana

The first major West African state in this era was the kingdom of Wagadu/ Ghana, which emerged in the Sahel - the semi-arid region between desert and savannah - from villages established during the period 500 to 700 ce. The principal people of Wagadu/Ghana were the Soninke, who were the northernmost of the Mande peoples, a large family of culturally related groups that would subsequently populate the Mali Empire. Some ancestors of the Soninke were probably among the Neolithic farmers who began cultivating sorghum and millet in the Sahelian grasslands during the period 3000 to 1000 bce. By about 1000 bce the Soninke ancestors began establishing small settled communities, and around 600 bce these grew into large villages

Map 22.i. Ghana/Mali/Songhay

administered by chieftaincies. The early farmers were among the first to take advantage of the iron technology that developed in West Africa by about 500 to 400 bce. As the most northern of the Mande peoples, the Soninke were also in contact with the nomads of the Sahara, from whom they acquired small horses brought from North Africa. The early Soninke's superior iron weaponry and horses made it possible for them to develop some level of a nascent polity. They gradually expanded their territories and dominated neighboring rulers until, by the tenth century, they had estab­lished a state (see Map 22.1).

Was Wagadu/Ghana a kingdom, an empire, or something else? Some have questioned the use of the term “empire” to describe ancient Ghana because it tends to obscure what scarce information is available in the sources. When applied to African polities like Wagadu/Ghana, questions of what is actually meant by such terms as “state," “kingdom,” and “empire” have generated a steady flow of theories by anthropologists, historians and geographers about African political systems in colonial times, through early post-colonial colloquiums on topics like the origin and development of state systems and into the present era of discussion about urbanization and state formation. West African scholars with the insider's perspective^[671] have contrib­uted valuable alternative theories to outsider's notions of “pristine states” versus “secondary states” or the ownership and application of technological developments. The amount of theoretical literature on these complex topics vis-a-vis West Africa is indeed extensive, even without taking into account evidence from the archaeology of pre-colonial cities and states. Considerable archaeological progress has been made in recent decades, but even so, Susan McIntosh concludes that given the present condition of the evidence, “any discussion of the development and political organization of early Sudanic polities must be both theoretical and speculative.”^[672] An example of current informed archaeological theorizing in this regard, would be the model of Wagadu/Ghana suggested by Roderick McIntosh, to the effect that it was not so much a despotic conquest state, as it was a “slowly consolidated confederation of many ^,chiefdoms^, in various relations to the core (from nominal, tribute-paying parity to fully administered).” McIntosh explains that his reconstruction is in line with recent views regarding “origins of states out of an earlier landscape of small-scale polities.” Regarding the later Mali and Songhay empires, he observes that “during times of dynastic struggles or other sources of political turmoil at the core, the elements of the periphery slipped easily into states of greater or lesser political autonomy.” Moreover, says McIntosh, during the rise of Wagadu/Ghana, “parts of the Middle Niger, such as the Jenne region, probably maintained their status as fully self-administering, independent, and friendly polities.”^[673]

The Soninke people's own ideas about their history are expressed in the “Legend of Wagadu,” an oral tradition told by many generations of gesere, professional oral historians and musicians of the Soninke. Details vary from one version to the next, but it generally describes the origins and early deeds of different Soninke clans. The story often begins by describing how the ancestor Dinga came from somewhere in the Middle East and settled in successive locations in what is today's Mali. In the Middle Niger Delta region he was at Jenne for a time, then at Dia, where he produced offspring who became Soninke ancestors elsewhere in the Sahel. Dinga is said to have eventually arrived at a place southwest of Nioro that was dominated by spirits of the bush, which in the story are called “genies.” Various versions describe a kind of magician's duel between Dinga and the genies with Dinga emerging victorious and marrying the three daughters of the chief genie. The sons from these wives became the ancestors of many Soninke clans, one of which was that of the Cisse. It was the Cisse clan that became the ruling lineage of Wagadu.

The next episode in the legend is reminiscent of the story of Jacob and Esau in the Bible. Dinga has grown old and blind, and before he dies he wants to pass his chiefly power on to his elder son Khine. But a younger son named Diabe disguises himself as the eldest brother and deceives his father into bequeathing him the chiefly powers. After Dinga dies, Diabe flees the wrath of his enraged elder brother and takes refuge in the wilderness where a mysterious drum falls out of a tree and lands at his feet. At the sound of the drum, four troops of cavalry come out from the four corners of the wilder­ness. The four commanders recognize Diabe as their superior and become his lieutenants. Later, after the Wagadu kingdom is established, they become chiefs or governors of the four provinces.

Diabe sets out to find a location where he can settle, arriving at a place called Kumbi, which was located in the southern part of what is now Mauritania, just north of the border presently shared with Mali. Arriving at the site where the town of Kumbi is to be established, Diabe finds it guarded by a giant snake called Bida. Significantly, the great snake is usually described as a python, implying the presence of water at the new settlement's location. A pact is made allowing Diabe to settle there with Bida as guardian, on the condition that every year a young virgin girl would be sacrificed to the great serpent. In return, Bida guarantees abundant rain for Wagadu and a plentiful supply of gold.

With its capital at Kumbi, Wagadu prospers under the rule of Diabe Cisse and his descendants known by the title Maghan. The descendants of Diabe and the four Fado or commanders of the provinces are still recognized as aristocratic Soninke clans called Wago. That term is probably related to “Wagadu” which is a contraction of wagadugu, a word that can be translated as “land of the Wago. ”

According to the legend, representatives of the four provinces of Wagadu would assemble at Kumbi each year to participate in the virgin sacrifice to Bida, the guardian serpent. This ceremony was the annual renewal of the pact between Diabe Cisse and Bida, to ensure the continued supply of gold and abundant rainfall. According to some versions, each year a different province was required to supply the virgin for the sacrifice. If this was the actual practice, it was a custom that probably helped maintain unity in the kingdom.

After an unspecified number of generations have passed, a year arrives when the virgin to be sacrificed happens to be the fiancee of a young man of noble birth. As the sacrifice is about to be made, the young man decapitates the snake with his sword, thus triggering the destruction of Wagadu. As Bida's severed head bounds into the sky it pronounces a dreadful curse that henceforth no rain would fall on Wagadu and no more gold would be found there. Deprived of rain and gold, Wagadu declines into ruin, its Soninke people are dispersed, and the countryside becomes a desert.

The Wagadu legend's mythical elements are obvious, but parts of it reflect both social and environmental realities that could have actually been involved in Soninke history. The kind of competition seen between the younger brother Diabe Cisse and his elder brother Khine, is a common form of stepbrother rivalry called fadenya in Manding society of which the Soninke are a part. At the state level of the early Western Sudanic kingdoms, there are many stories of brothers being involved in bloody rivalries for succession to the throne (especially in the Songhay Empire).

With regard to environmental elements in the legend, it is a fact that pythons are equally at home in the water and on land. Their presence was a sure sign of a climate with adequate wet phases to support a settlement, as suggested by the bargain struck between Bida and Diabe Cisse. In early times before the arrival of Islam and Christianity, the great pythons were sacred religious symbols throughout sub-Saharan West Africa from the Sahel to the Atlantic coast. In recent times, zoologists have found that during the heat of the day in the dry season, pythons usually seek water in which to submerge themselves. Thus, it is not difficult to see how the idea of the great snake as a highly spiritual water oracle could develop.

For the early Arab geographers who wrote about “Ghana,” it was a tantalizing land of mystery and fabled wealth. They had some inaccurate and even fantastic notions of what was there, but they also provide important information. In 738 a governor of the Maghrib sent a commercial expedition to “the land of the blacks,” and the merchants returned laden with slaves and gold. The trade seems to have originated with the peoples of the desert, the most powerful of whom were the Sanhaja who carried on trade with the Soninke to the south of them. The Soninkc's early involvement with the traders of the Sahara is one reason that Wagadu/Ghana emerged as a great power ahead of other chiefdoms of the medieval Sahel. The other main reason is that Wagadu controlled the sources of both salt and gold. As Ibn Hawqal, writing in 988, described it, “the ruler of Ghana is the wealthiest king on the face of the earth because of his treasures and stocks of gold extracted in olden times for his predecessors and himself.”^[674]

Efficient food production, early control of iron technology for superior weaponry and the acquisition of horses, helped the Soninke achieve early superiority over their neighbors. The Muslim writer al-Ya‘qubi (d. 897) described Ghana as one of the two most powerful kingdoms of the Western Sudan, with a ruler who had other kings under his authority. What eventu­ally raised the Soninke kingdom to the imperial level was its control of both regional and trans-Saharan trade. The inter-regional trade involved the exchange of salt, copper and dates from the Sahara. Savannah products included slaves, livestock, iron tools, weapons and utensils, animal hides, leather goods such as sandals, cushions and bags, locally woven and dyed cloth, clay pottery, woven grass products like baskets and sleeping mats, medicinal herbs, and foodstuffs like dried fish, rice, various grains and condiments, spices, honey, and fruit. From farther south nearer the forest came gold and kola nuts.

On a broader scale, Wagadu/Ghana was well positioned to dominate the international caravan trade across the western Sahara. One of the most important reasons for that commercial development had been the introduc­tion of the camel into North Africa by the Romans during the first century ce. The one-humped camel was originally domesticated in southern Arabia around 5000 bce, and introduced into northeastern Africa around 3000 bce. From northeastern Africa the camel arrived in the Sahara Desert sometime during the first centuries ce.It was from that time on, thanks to the camel, that regular and extensive trade across the Sahara became possible.

In the second and third centuries ce the use of camels expanded among North African Berber peoples. The Sanhaja Berbers acquired significant numbers of camels by the fourth and fifth centuries, and began to develop and control increasingly busy desert trade routes. The trans-Saharan caravans could consist of as few as a half-dozen camels, or as many as two thousand. They usually left North Africa in April or May, led by professional Berber guides who could find the wells and waterholes. The dangerous journey lasted from two and a half to three months, depending on the size of the caravan and conditions of the route. Unusually dry years could leave the wells with insufficient water, and if a severe sandstorm came up the entire caravan could perish.

In pre-Islamic times, the Maghrib and southern Sahara were already linked by Berber-speaking nomads who interacted with Sahelian sedentary popula­tions of the Bilad al-Sudan (“land of the blacks”). Sometime during the eighth century the Zanata and other Berber peoples of the Atlas region became Muslims, but it was not until the first half of the eleventh century that the Sanhaja Berbers of the southern Sahara became committed to Islam when they were coerced into joining the Almoravid movement. Extensive conver­sion of the Berber peoples brought with it wider commercial connections, increased the scale and complexity of their trade, and generally enhanced their prosperity. This was evident in sub-Saharan commercial centers includ­ing the much-contested city of Awdaghust, which originally gained commer­cial prominence under Berber authority. During the early decades of the eleventh century, the Soninke of Wagadu/Ghana seized control of Awdaghust, but Almoravid Sanhaja retook it in 1055-6.

By the eleventh century, black sub-Saharan rulers of Wagadu/Ghana in the western Sahel and Gao on the Niger Bend were accepting the presence of North African Muslim traders in designated quarters of their cities while most of the local population continued with their traditional spiritual prac­tices. For merchants, participation in the vast Sahelian commercial network became dependent on at least nominal conversion to Islam, even for black sub-Saharan traders that included Fula, Wolof and Manding. It was the latter who linked the savannah goldfields of Bambuk and Bure with the sub­Saharan entrepots of Walata, Jenne, and Timbuktu.

Wagadu/Ghana's advantageous location in the Sahel enabled the Soninke to function as middlemen controlling commerce from the savannah and forest zones in the south, and the Sahara and Maghrib in the north. From tributary states in the Senegal Valley the Tegdaoust/Awdaghust entrepot received salt, gold, ivory, dried/salted fish, produce, and craft products. From the broader savannah region and the forest it received gold, elephant and hippopotamus ivory, ebony, slaves, ostrich feathers, wild and domestic animal hides, gum arabic, produce such as dates and kola nuts, pottery, leatherwork, and other craft goods. The geographer Yaqut (1179-1229), a freed slave of Greek origin who became a Muslim, described the country's commercial position: “Merchants meet in Ghana and from there one enters the arid wastes towards the land of Gold. Were it not for Ghana, this journey would be impossible, because the land of Gold is in a place isolated from the west in the land of the Sudan. From Ghana the merchants take provisions on the way to the land of Gold.”^[675]

The north-south trade employed a network of routes connecting Waga- du's TegdaoustZAwdaghust entrepot with its counterpart trading city of Sijilmasa in the Maghrib, as well as with al-Andalus, Tripoli and Egypt. From those sources came manufactured objects and luxury goods from the Medi­terranean world, Europe, and North Africa. They included iron products like knives, scissors, needles, and razors; brass and copperware; textiles including silk, velvet, and brocade; ornaments and jewelry, including glass and porcel­ain beads; and other luxury goods, including silver, glassware, mirrors, carpets, perfumes, paper, tea, coffee, and sugar. Horses from North Africa were one of the most important items moving south, as were cowrie shells, which were used as currency in West African markets.

As the twelfth century unfolded, the dominance of WagaduZGhana grad­ually faded away. Desert encroachment on formerly productive land as well as generations of struggle with powerful desert nomads prompted many Soninke to abandon their ancestral lands and move to a less difficult and capricious environment. The decline of Soninke power left a vacuum that was filled for a time by some smaller savannah kingdoms to the south in areas closer to rivers and lakes, and with more reliable rainfall. In the first half of the thirteenth century some Mande chiefdoms (jamanaw) of the Upper Niger began to unify into a new state that would eventually develop into the Mali Empire.

The Mali Empire

All of the above-mentioned goods that were traded in the markets of WagaduZGhana from the tenth to twelfth centuries continued to generate revenue in the markets of Mali from the thirteenth to fifteenth centuries, but gold was the most important commodity. There were three principal gold­fields below the Sahara. One of the main ones, which had also been a source for WagaduZGhana, was at Bambuk, between the Senegal and Faleme rivers. Another, also formerly controlled by Wagadu/Ghana, was at Bure above the Upper Niger in what is now northeast Guinea. The third was in Akan territory near the forest in the modern republics of Cote d'Ivoire and Ghana. Mali drew on all three goldfields for the trans-Saharan trade in precious metal for which merchants from North Africa, the Middle East, and Europe competed. The exact routes or frequency of shipment from the individual goldfields are not entirely known, but much of the gold from Bure was moved along the Niger River to Jenne, where camel caravans headed overland across the Sahara, with some continuing to Timbuktu before turning north.

The Mande people's oral tradition, “Mande Maana,” which presents their own perceptions of the origins of the Mali Empire, is usually referred to outside the culture as the “Sunjata Epic.” In recent decades the corpus of indigenous oral sources has grown enormously, and many previously unknown variants of the Sunjata epic, among other important narratives, have come to light. The additional oral sources have been accompanied by increasingly sophisticated methodologies for determining reasonable degrees of probability for elusive elements of historical evidence.

The Sunjata epic is named for Sunjata Keita, who, along with several other charismatic ancestors of both genders, is credited with a major role in establishing the foundations of the Mali Empire. The story begins sometime around the beginning of the thirteenth century in Farakoro, a Mande chief- dom. The identification of Farakoro and other places as important locations in Mali's political history is relatively recent. From the early 1960s when D. T. Niane introduced the Sunjata Epic to the world outside West Africa, the town of Niani on the Guinean side of the Sankaran River was said to be the home of Sunjata's father and the “capital” of the Mali Empire. More recently, a significant amount of evidence has been found to indicate that Konfara was the chiefdom of Sunjata's father and his village was Farakoro. This was near the goldfields of Bure, which had been one of the main sources of gold for Ghana in earlier centuries and proved to be similarly important for the Mali Empire. One of Sunjata's most important towns was Dakajalan, said to be the place where he spent his early years, to which he returned from exile, and which served as his headquarters during the war with Soso.

Like all powerful men of Mande culture, Sunjata's father, Maghan Konfara, had diviners who would forecast the future. They told him that he would be the father of a great hero, but that the woman who would be the hero's mother had not yet been found. After a long search the woman is finally located in the kingdom of Do ni Kiri. She is Sogolon Conde, a sister of the mansa (king) of Do ni Kiri. Sogolon is an ugly, hunchbacked woman with formidable powers as a sorceress and is recognized as the woman who is destined to give birth to Sunjata. She is brought to Farakoro and married to Maghan Konfara, who already has many other wives. Jealous of the diviners' prediction favoring Sogolon, the co-wives do everything they can to stop her from giving birth to the hero. After several years of trouble the predicted birth takes place, but the child is born crippled. He is called “Sogolon's Jara” (jara = lion). It takes years for Sogolon's Jara (Sunjata) to learn to walk, but when he finally does he becomes a great hunter. One of Maghan Konfara's other wives has a son who was born before Sunjata. Despite the diviners' forecast of greatness for Sunjata, she is determined that her son will be the next chief. After a failed attempt to kill Sunjata, Sogolon moves him and her other children into exile, eventually settling in the old Soninke kingdom of Mema in the Middle Niger Delta.

While Sogolon and her children are in exile, the Mande chiefdoms are conquered by the army of Soso led by its powerful king Sumaworo Kante. After long suffering under his tyrannical rule, the Mande people recall the diviners' prediction that Maghan Konfara's wife Sogolon would give birth to a great leader. A search party locates Sunjata in Mema and returns him to Manden where he organizes the Mande chiefdoms into a powerful army against Soso tyranny. Sunjata's army eventually vanquishes the army of Soso, and the unified Mande chiefdoms form the basis of a powerful kingdom that expands into neighboring territories and becomes the Mali Empire. The Mande oral traditions do not give dates for the events they describe, but according to the Arabic sources, Mande's defeat of Soso happened sometime during the 1230s. The most detailed information about Mali is provided by three Arab geographers and historians: Al-^tUmari (1301-1349), Ibn Battuta (1304-1368 /9 or 1377) and Ibn Khaldun (1332-1406). Of these, only Ibn Battuta actually visited the court of Mali.

Ibn Khaldun heard that Mali became the greatest power in the Western Sudan. He reported that the greatest king of Mali, who overcame the Soso and conquered their country was named Mari Jata (a.k.a. Sunjata), who ruled for twenty-five years. Sunjata was succeeded by his son Mansa Wali (mansa = “king” or “emperor”) who survived in memory because he made a pilgrim­age to Mecca during the reign of the Mamluk Sultan Baybars, who ruled Egypt and Syria (1260-77). The relatively sparse information available about rulers other than a famous few indicates that Mali had major leadership problems at various times, evidently suffering the fatal flaw of never estab­lishing an effective standard of royal succession. In the decades between Sunjata's reign and those of Mansa Musa and Mansa Sulayman in the fourteenth century, the country endured repeated periods of dynastic instability, a pattern that would be repeated after the latter's death. Failed rulers ranged from the merely undistinguished to the insanely homicidal. They included a vicious tyrant, an assassin, a brazen plunderer of state wealth, some well-intended but ineffective weaklings, and usurpers who stepped in at times with mixed results. The Arabic name of Mansa Abu Bakr reflects the continuing influence of Islam among the ruling elite, and the reign of a formerly servile military commander named Sakura (1298-1308) underlines the periodic failures of some descendants of Sogolon Conde's offspring, Sunjata and Manden Bori. Upon Sakura's death the kingship reverted to two forgettable descendants of Sunjata before power passed to descendants of Manden Bori. This finally resulted in the reign of one of Mali's great rulers, Mansa Musa (r. 1312-37).

By this time the desert city of Walata had overtaken Tegdaoust/Awda- ghust as the main western terminus of trans-Saharan trade. However, Jenne was linked to Timbuktu by the Niger River, and it subsequently became the key distribution center for salt moving southward from the central Sahara via Timbuktu and, among other products, gold transported north to the Magh­rib. At the regional level, the main difference between Wagadu/Ghana and Mali had been that at the height of Mali's power it controlled far more territory than Ghana ever did, so it had more resources to exploit. By the beginning of the fourteenth century, Mali's expansion into the Inland Delta, Gao, and what would eventually become eastern provinces of Songhay, had added enormously to the farming, grazing, hunting and fishing resources of the empire. The new territories also provided additional sources of slaves for trade, military service and farm production. Tribute from newly subordin­ated kings and chiefs, and tariffs from newly controlled trade routes enriched the government treasury.

Thus, Mansa Musa's twenty-five-year reign is thought of as the “golden age” of Mali. Among sub-Saharan West African rulers who made the pil­grimage to Mecca, Mansa Musa was the most famous. His caravan is said to have included eighty loads of gold dust, and his 1324 arrival in Egypt created a sensation because Mansa Musa distributed so much gold as gifts, and the Malians spent such large amounts in the market, that it took several years for the value of gold to recover (see Figure 22.1). Publicity resulting from the sensational visit brought increased awareness of Mali to North Africa, the Middle East, and Europe. An impression spread throughout the Mediter­ranean world of vast wealth to be had south of the Sahara. Trade with the

Figure 22.ι Illustration of Mansa Musa in detail from the Catalan Atlas, 1375 (vellum), Cresques Abraham (1325-87) (Bibliotheque Nationale, Paris, France / Bridgeman Images)

Maghrib was especially boosted by Mansa Musa's exchange of diplomatic embassies with the sultan of Morocco. Increasing numbers of traders from throughout North Africa embarked on commercial ventures across the Sahara. By the mid-fourteenth century these included Egyptian traders making regular visits to Mali. Citizens in commercial centers like Walata were dressing in clothes and buying other imported goods from Mediterra­nean shores.

Another especially significant result of Mansa Musa's pilgrimage was the arrival in Mali of a Muslim architect recruited from Spain named Abu Ishaq al- Sahili. The architect built Mansa Musa a rectangular domed house covered with plaster that was decorated with colorful designs, thus introducing an architectural style that can still be seen in many towns and cities of the Western Sudan. One of Mansa Musa's residences was in Timbuktu, location of the still- famous Sankore Mosque which is believed to have been designed by al-Sahili.

When Mansa Musa died in 1337, his son Mansa Magha succeeded him, but was replaced after a few years by Musa's brother, Mansa Sulayman (r. 1341-60). Although not popular with his subjects, Sulayman was a power­ful and effective ruler of the empire. In 1352-3 his court was visited by the North African geographer Ibn Battuta who later wrote an eyewitness account indicating that the royal court of Mali was as rich and splendid as any in the Muslim world. The palace throne room is described as a “lofty pavilion” with curtained, gilded arches and silken accessories. An upholstered outdoor throne was shaded by a large silken parasol topped by a golden falcon. Royal audiences featured elaborate pageantry and ceremony, with marching drum­mers and trumpeters with elephant tusk horns, and hundreds of armed guards formed in ranks. Ibn Battuta describes the pomp and circumstance of the mansa’s ceremonial approach to the throne wearing a golden head­dress and red robe, carrying a bow and quiver of arrows, and preceded by singers and musicians “with gold and silver stringed instruments.”^[676] Never­theless, grand as Ibn Battuta found the royal court to be, the pious traveler was deeply offended by the casual social relationships he saw between men and women. In the royal compound, he witnessed the public nudity of scores of female slaves and servants, which Jerry Bentley points to as clearly indicating that “pagan traditions” survived there, with Islamic law being established in only limited and selective fashion.^[677]

The reigns of Musa and Sulayman spanned the years 1312-60 and marked the apogee of the Mali Empire. When Mansa Sulayman died in 1360, his sons and the sons of Mansa Musa fought over the succession. Civil war ensued, opening the way to generations of power struggles and leadership gone awry. By the end of the fourteenth century, generations of power struggles and weak leadership had undermined Mali's power to the point where it became impossible to retain control of the empire's distant frontiers; indeed Mali lost control of Timbuktu sometime around 1433. Beyond the Niger Bend, the more distant eastern provinces, including Gao, had probably been lost even earlier.

The kingdom of Gao

Among the early people of the Niger Bend region were the camel-riding Sanhaja of the Sahara Desert. Locally known as Tuareg, they rode out of the great desert to establish trading camps near the Niger River. Other early settlers were a riverine people known as Sorko, who moved northward into the region from an area known as Dendi, which was downstream from Gao. Sorko origins are obscure, but they were early speakers of proto-Songhay, either bringing it with them or adopting it later under the rule of Songhay horsemen who followed them upriver from Dendi and established control over them. Between 750 and 950, while Ancient Ghana was prospering as “the land of gold” far to the west, the trading center at Gao became an increasingly important southern terminus for trade across the Sahara Desert. By the tenth century the rulers of Gao had established it as a small kingdom, taking control of the peoples that lived along the trade routes. By around 1300, Gao had become so prosperous that it attracted the attention of Mali's rulers and, as we saw earlier, was conquered by them. Until recently, modern scholarship on the history of the Songhay polity relied largely on chronicles written in Timbuktu during the second half of the seventeenth century. According to these chronicles, the Songhay state was ruled by a series of three dynasties. They began with rulers who carried the title Zuwa, the first fourteen of which ruled before the arrival of Islam. The second dynasty was that of the Sii or Sonyi (c. 1430-1492), and they were followed by the Askias who held power from 1493 until the Moroccan invasion of 1591.

Linguistic and other evidence suggests a fair likelihood that the rulers known as Sii or Sonyi, originated as a Mande warrior group that moved in from Mali. According to the now familiar narrative presented in the Timbuktu chronicles and supplemented by oral tradition, the Sii dynasty was highlighted by the reign of Sii Ali Beeri (“The Great”), who transformed the kingdom of Gao into the empire of Songhay, which would replace Mali as the great power of the Western Sudan. Sii Ali commanded a large, well- disciplined army and a fleet of riverboats with which he expanded Gao's territory throughout the Middle Niger Delta. Ali's control of the Niger Bend brought with it the rich gold and salt trade that passed through Timbuktu and Jenne, which became the second and third most important cities of the Songhay state.

According to the Timbuktu chronicles, the Sii dynasty was followed by one that adopted the title Askia, which has long been said to date from the 1480s,^[678] although epigraphic evidence uncovered by Paulo Moraes Farias reveals that the title was in use in Gao by 1234, two and a half centuries earlier. Under the first Askia, who became known as Askia Muhammad the Great (1493-1529), Songhay would flourish into the early decades of the sixteenth century. Imperial control eventually expanded northward to the saltpans of Taghaza in the Sahara Desert, westward to former territories of the Mali Empire, and eastward to the Tuareg sultanate of Agadez. The empire became so large that its army was divided into two, one for the western provinces based in Timbuktu, and one for the eastern provinces based in Gao. The sons of Askia Muhammad were mostly half-brothers, and in 1529 Askia Muhammad was deposed by one of them. Similar to the case of Mali, failure to establish an effective process of succession would eventually lead to Songhay’s downfall. Askia Muhammad's son Musa (1529-31) had no compunctions about killing rival brothers to stay in power. This set the tone for the triumphs and vicissitudes of the seven succeeding Askias up to the time of the Moroccan conquest of Songhay in 1591.

With the publication of Arabic Medieval Inscriptions from the Republic of Mali in 2003, Paulo Moraes Farias has introduced epigraphic evidence revealing that the Timbuktu chronicles were not the product of detailed historical records extending back over several centuries as had been previously thought, but were constructed on ideological grounds. He describes their production on one level as being “an exercise in catastrophe management,” catering to emotional and intellectual needs that resulted from the destruc­tion of the Songhay Empire by the Moroccan invasion of 1591. Moraes Farias reasons that through artful reconstruction of the past the chroniclers were coming to terms with the disastrous events that reduced the Askias to puppet roles and caused personal suffering for the literate urban elite and down­graded their political and social status. He argues, moreover, that on another level the Timbuktu chroniclers were endorsing acceptance of the post­invasion Arma (Arabic: al-rumdh - “musketeers”) regime while at the same time formulating grounds for its reform. Thus, the chroniclers’ political agenda required consolidation of the Songhay and Timbuktu past into a seamless narrative with chronological depth and political continuity. Moraes Farias argues that the chroniclers, in pursuit of their aim to represent the Songhay state as an unbroken narrative spanning three successive dynasties, manipulated their lengthy king lists, obfuscating periods of dynastic change and interval. The epigraphic evidence has convinced Moraes Farias that the Timbuktu chronicles artificially shorten Mali’s period of domination of the eastern arc of the Niger Bend, and that they also prolong the length of the Zuwa (or Za) dynasty into the past and possibly forward in time.

In the epigraphic evidence there are four identifiable series of royal titleholders, with two of the series of rulers known as Malik (pl. Muluk). The earlier Muluk (kings) held power from an undetermined date up to 1083-4,^[679] and the later series of Muluk reigned from c. 1084 to c. 1203. Names of ancestors appear in the later Muluk series of inscriptions, but without explicit reference to royal titles, which leads Moraes Farias to the tentative conclusion that the series of rulers known as Muluk was based on a system of circulating kingship, which was apparently more important to them than claims to royal ancestry.

The third series of rulers featured women known by the title Malikat (queen) which Moraes Farias believes existed parallel to, but relatively autonomous from, the Muluk series. The last-known inscription date for the later Muluk series is 1203 and the earliest date for the officeholders called Zuwa is 1127, so there is a chronological overlap of the two. Moraes Farias qualifies as “provisional” his description of the Zu'a/Zuwa series as a succession of kings, because it is not possible to be sure of the exact nature of their politico-ritual role, especially in regard to the earliest of them. Moreover, the title Malik is explicitly attached to only two of the inscriptions. Among possible explanations, Moraes Farias hypothesizes that during the twelfth century, political power over Gao and neighboring areas might not have been unified, with some communities governed by different rulers who coexisted peacefully. Another possibility is that twelfth-century Zuwa served under the Muluk in offices that did not challenge their authority, and then replaced the Muluk with overall power after 1203.

The last inscription date found for the Zuwa officeholders can be read as either 1280 or 1299, and this appears to mark the end of independence for Gao and the beginning of Mali's rule over the Middle Niger.^[680]

Other states

Although Wagadu/Ghana, Mali, and the Gao/Songhay state tend to receive the most attention in the historical literature, they were by no means the only significant polities in medieval West Africa. In the ninth to the eleventh centuries, the Kingdom of Takrur was the dominant power in the Senegal River Valley, competing for trade with Wagadu/Ghana as a zealously Islamic ally of the rising Almoravid movement of the western Sahara. The Almora- vid movement commenced in the second half of the tenth century when ‘Abdallah ibn Yasin (d. 1059), a Berber of the desert-dwelling Sanhaja people, began to establish what would become a formidable Islamic empire which, as we saw earlier, rose to power during the decline of Wagadu/Ghana. The Almoravids took control of the Soninke territories on the way to achieving political hegemony for the next two centuries throughout the western Sahara and the Maghrib, where they founded the city of Marrakesh in 1062. At its apogee, the Almoravid Empire included present-day Mauritania, western Sahara, Morocco, western Algeria, and southern Spain.

Some relatively ephemeral kingdoms of the period were, for a time, client states or provinces of Wagadu/Ghana. These chiefdoms or smaller states were generally much slower to adopt Islam, in most cases incorporating Muslim elements into traditional belief systems that constituted the indis­pensable underpinnings of power and authority for local rulers. Such was the case with Kaniaga and Diafunu, each of which briefly occupied a Sahelian power vacuum left by the decline of Wagadu/Ghana, as well as Kingui and Mema, which were located west of Gao in the Middle Niger Delta and later absorbed by the Mali Empire. But the most powerful state in the interim between the decline of Wagadu/Ghana and the rise of Mali was the King­dom of Soso, whose imperial ambitions ended in the first half of the thirteenth century with the alliance of rival Mande chiefdoms that conquered Soso and established the foundations of the Mali Empire.

Conclusion

In the period between 500 and 1500 ce, West African settlements between the southern edge of the Sahara Desert and major riverine environments, including Senegal and Niger, experienced varying degrees of success in taking advantage of their geographical locations and control of natural resources to establish and sustain urban centers and satellite communities that eventually emerged as polities of the Western Sudan. Key values held in common were elaborate kinship systems, customs and rituals addressing issues of inter-ethnic relations, and spiritually based constructions of power and authority. These archetypal core values of Mande society made it possible for ancient urban centers to emerge by themselves and endure for more than a millennium and a half. In place of despotic rule as found in later- developing hierarchical state systems like Wagadu, Mali, and Songhay, urban centers passed through multiple stages of development in a process that culminated by the beginning of the Middle Millennium in an urban complex that was innovative and flexible enough to achieve long-term sustainability. In the earliest instances, prior to the emergence of ruling hierarchies, these were governed through heterarchy, i.e. a network of groups of competing, overlapping interests including farmers, herders, fishermen, hunters, artisans, and merchants.

Among the Western Sudanic polities that subsequently developed under hierarchical authority, Wagadu/Ghana was one of the earliest to experience the benefits of the trans-Saharan trade and expand into neighboring territories. Well placed to function in a middleman capacity, Wagadu/Ghana gained control of sources of artisanal, agricultural, and mineral goods, commanding the gold-producing center of Bambuk and, by the eleventh century imposing control over the important Berber trading center of Awdaghust in the south­western Sahara. The eleventh-century rise of the Almoravids and the decline of Wagadu/Ghana from the late twelfth century resulted in the use of trade routes farther east in what became the heartland of the Mali Empire. By the fourteenth century, as Mali became the dominant Western Sudanic power, the Bure goldfields, near tributaries of the Upper Niger in present-day Guinea, came into play, with Jenne and Timbuktu emerging as important trading centers, all under Malian control. Meanwhile, farther east and sometime before the tenth century, the arc of the Niger Bend had been settled by hunters, fishermen, boatmen, farmers, herders, and artisans. They were subsequently joined by horse-riding Songhay speakers who brought both their language and political dominance to the region. During the first half of the fifteenth century, the kingdom of Gao emerged, prospered, and began filling a power vacuum left as the Mali Empire slid into decline. Gao subsequently expanded into the third of the great empires of medieval West Africa, with both people and state known as Songhay. Established in 1493, the last ruling dynasty (the Askias) of the Songhay Empire prevailed until the Moroccan invasion of 1591 brought to a close the era of the great medieval states of the Western Sudan.

FURTHER READING

Brooks, George. Landlords and Strangers: Ecology, Society, and Trade in Western Africa, 1000-1630. Boulder, CO: Westview Press, 1993.

Cisse, YoussoufTata and Wa Kamissoko. La grande geste du Mali des origines a Iafondation de !’Empire. Paris: Karthala-Arsan, 1988.

Soundjata, la gloire du Mali. Paris: Karthala-Arsan, 1991.

Connah, Graham. African Civilizations: Precolonial Cities and States in Tropical Africa: An Archaeological Perspective. Cambridge University Press, 1987.

Conrad, David C. “From the Banan Tree of Kouroussa: Mapping the Landscape in Mande Traditional History,” Canadian Journal of African Studies 42 (2008): 384-408.

“Mooning Armies and Mothering Heroes: Female Power in Mande Epic Tradition,” in Ralph Austen (ed.), In Search of Sunjata: The Mande Epic as History, Literature and Performance. Bloomington, IN: Indiana University Press, 1999: 189-229.

“Oral Tradition and Perceptions of History from the Manding Peoples of West Africa,” in Emmanuel Kwaku Akeampong (ed.), Themes in West Africa's History. Athens, OH: Ohio University Press, 2006: 73-96.

“A Town Called Dakajalan: The Sunjata Tradition and the Question of Ancient Mali's Capital,” Journal of African History 35 (1994): 355-77.

Conrad, David C. and H. J. Fisher. “The Conquest that Never Was: Ghana and the Almoravids, 1076,” History in Africa 10 (1983): 53-7.

Garrard, Timothy. “Myth and Metrology: The Early Trans-Saharan Gold Trade,” Journal of African History 23 (1982): 443-61.

Hunwick, John, trans. and ed. Timbuktu & the Songhay Empire: Al-Sa'di's Ta'rikh al-sudan down to 1613 and Other Contemporary Documents. Leiden: Brill, 1999.

Kea, Ray A. “Expansions and Contractions: World-Historical Change and the Western Sudan World-System (1200/1000 b.c. - 1200/1250 a.d.),” Journal of World-Systems Research 10 (2004): 723-816.

Law, Robin. The Horse in West African History. Oxford University Press, 1980.

Levtzion, Nehemia. Ancient Ghana and Mali. London: Methuen, 1973.

“The Western Maghrib and Sudan,” in The Cambridge History of Africa, vol. ιιι. Cambridge University Press, 1977: 331-462.

Levtzion, Nehemia and Randall L. Pouwels, eds. The History of Islam in Africa. Athens, OH: Ohio University Press, 2000.

McDougall, E. Ann. “The Sahara Reconsidered: Pastoralism, Politics, and Salt from the Ninth through the Twelfth Centuries,” African Economic History, 12 (1983): 263-86.

“The View from Awdaghust: War, Trade and Social Change in the Southwestern Sahara, from the Eighth to the Fifteenth Century,” Journal of African History 26 (1985): 1-31.

McIntosh, Roderick J., Joseph A. Tainter, and Susan Keech McIntosh, eds. The Way the Wind Blows: Climate, History, and Human Action. New York, NY: Columbia University Press, 2000.

McIntosh, Susan Keech. “Floodplains and the Development of Complex Society: Com­parative Perspectives from the West-African Semi-Arid Tropics,” in Elizabeth A. Bacus and Lisa J. Lucero (eds.), Archaeological Papers of the American Anthropological Association 9 (1999): 151-65.

“A Reconsideration of Wangara/Palolus, Island of Gold,” Journal of African History 22 (1981): 145-58.

Mitchell, Peter. African Connections: Archaeological Perspectives on Africa and the Wider World. Walnut Creek, CA: Altamira Press, 2005.

Moraes Farias, P. F. de. “Intellectual Innovation and Reinvention of the Sahel: the Seventeenth-Century Timbuktu Chronicles,” in Shamil Jeppie and Souleymane Bachir Diagne (eds.), The Meanings of Timbuktu. Cape Town: HSRC Press, 2008: 95-107.

Munson, P. J. “Archaeology and the Prehistoric Origins of the Ghana Empire,” Journal of African History 21(1980): 457-66.

Niane, Djibril Tamsir. Sundiata: An Epic of Old Mali. Trans. G. D. Pickett, London: Longman, 1965.

Nicholson, Sharon. “Saharan Climates in Historic Times,” in M. A. J. Williams and Hugues Faure (eds.), The Sahara and the Nile: Quaternary Environments and Prehistoric Occupation in Northern Africa. Rotterdam: A. A. Balkema, 1980: 173-200.

as-Sa'di, ‘Abd ar-Rahman b. ‘Abdullah. Ta'rikh as-Sudan, trans. and ed. Octave Houdas and Maurice Delafosse. Paris: Librarie d'Amerique et d'Orient Adrien-Maisonneuve, 1964.

Sutton, John E. G. “West African Metals and the Ancient Mediterranean,” Oxford Journal of Archaeology 2 (1983): 181-8.

Webb, James. Desert Frontier: Ecological and Economic Change Along the Western Sahel 1600-1850. Madison, WI: University of Wisconsin Press, 1995.

Willis, John Ralph. “Ancient Ghana and Mali,” International Journal of African Historical Studies 8 (1975): 175-81.