Wagadu and its king
The first great Sahel state was built by the Soninke people, who called it Wagadu. Arab writers called it Ghana, after a title of its king. It flourished from at least the eighth century until the late eleventh, though its roots are probably older. The earliest known Arabic reference, by the astronomer al-Fazari in the late eighth century, called Ghana "the land of gold." That reputation colored everything outsiders wrote about it.
Ghana did not, in fact, sit on top of the gold. The main gold fields lay at Bambuk, between the Senegal and Faleme rivers, beyond the kingdom's southern edge. The kings' power came from taxing the trade rather than owning the mines. Goods entering and leaving the kingdom paid duties, and merchants who wanted the gold had to deal with the court.
Two towns side by side
Writing in Muslim Spain in 1068 from merchants' reports, the geographer al-Bakri described the capital as two towns about six miles apart: a Muslim merchants' town with twelve mosques, and the king's town, surrounded by sacred groves where priests lived and where the kings were buried. The king followed the traditional religion, yet Muslims served at his court as ministers, interpreters and treasurers, because they could read and write Arabic and knew the merchants of the north. Royal audiences were grand. Al-Bakri describes the king in a domed pavilion, with horses in gold trappings, dogs wearing gold collars and pages bearing swords and shields.
Archaeology may let us walk those streets. Koumbi Saleh, in southeastern Mauritania, is widely though not universally identified as the merchants' town of Ghana's capital. Excavations there, beginning in 1914 and continuing at intervals into the 1970s, uncovered stone houses, a large mosque and cemeteries. The king's town has never been securely located, perhaps because it was built of earth and wood that left little trace.
The routes north
Caravans linked Ghana to Sijilmasa in southern Morocco, the great northern market for West African gold. On the southern edge of the desert sat Awdaghust, an oasis town of Berber merchants that at times paid tribute to Ghana. The crossing between them took many weeks. Gold from Ghana and its neighbors fed the mints of North Africa and Spain, and some of it reached Europe and the Middle East.
Arab writers also told of a strange custom at the edge of the gold lands. In what is often called "silent trade," merchants laid out goods, withdrew, and let unseen local people leave gold beside them, each side adding or taking until both were satisfied, without the two parties ever meeting. The Greek historian Herodotus had told a similar tale about Carthaginian traders centuries earlier. Historians suspect the story is at least partly a traveler's legend, though it may reflect real efforts by gold producers to keep their mines secret.
The serpent of Wagadu
Soninke oral tradition explains Ghana's fall in its own way. In the legend of Wagadu, the land prospered under the protection of a great serpent named Bida, which demanded a young woman each year in return for rain and gold. When a young man killed the serpent to save his bride, the serpent's curse brought drought, the gold vanished, and the Soninke scattered. Some historians hear in the story a memory of real change: the Sahel did grow drier over these centuries, and Soninke traders did disperse across the region.
Worked example: did the Almoravids conquer Ghana?
In the mid eleventh century a reform movement among the Sanhaja Berbers of the western Sahara, the Almoravids, swept north into Morocco and Spain. They also took Awdaghust in 1054 to 1055. Many textbooks once stated that they conquered Ghana itself in 1076 and forced its people to convert. What does the evidence show?
Al-Bakri, writing in 1068, describes Ghana as strong, with no hint of coming disaster. The Andalusian geographer al-Zuhri, writing in the twelfth century, says the people of Ghana became Muslim around 1076, but does not clearly describe a war. The fullest account of a conquest comes from Ibn Khaldun in the late fourteenth century, three hundred years after the event. In the early 1980s the historians David Conrad and Humphrey Fisher argued that the Almoravids never conquered Ghana by force, and that its rulers adopted Islam through peaceful contact.
Critics reply that silence in thin sources proves little, and that Soninke traditions do recall conflict with desert peoples. The fair conclusion is that the conquest is unproven. Ghana certainly weakened in the late eleventh and twelfth centuries, its rulers converted, and power shifted to new states. Whether this happened by the sword, by trade and preaching, or by a mix of all three remains open. The case shows how a single late source, repeated often enough, can harden into a fact that everyone knows and no one checks.
