CHAPTER 8

Ghâna, One Hundred
Years Later

Banks of a River in the Sahel, between 1116 and 1154

Around 1154, the year he began his famous map of the world for the Christian king Roger II of Sicily, the Arab geographer al-Idrîsî described “the greatest of all the towns of the Sûdân” (i.e., the “Blacks”), “the most populous, and with the most extensive trade.” “Prosperous merchants,” he wrote, “go there from all the surrounding countries and the other countries of al-Maghrib al-Aqṣā,”—the “most Maghreb” or the farthest west of Islam and the known world: present-day Morocco. The king had a palace here, “strongly built and perfectly fortified, [whose interior is] decorated with various drawings and paintings, and provided with glass windows.” The construction of this palace began, al-Idrîsî tells us, in the year 510 of the Hijra, that is, 1116–1117 CE, which furnishes the terminus post quem for the information he relates. As for al-Idrîsî’s informants, the Maghrebians who traded from one side of the Sahara to the other are the most likely candidates. According to the author, they “know positively and without difference of opinion that the king has in his palace a brick (libna) of gold weighing 30 raṭls [approximately 35 pounds or 16 kilograms] made of one piece. It is entirely God’s creation, without having been melted in the fire or hammered with any tool. A hole has been pierced in it to serve for tethering the king’s horse. It is one of those curious objects which no one else possesses and which is not permissible for anyone else. He prides himself on it above the other kings of the Sûdân.”

This city was called Ghâna. If all the merchants of North Africa knew it, it was certainly because it was the place in the Sahel where gold was purchased in the middle of the twelfth century. Moreover, the king did not dispense with the ostentatious displays of wealth that made it known: the anecdote about the reins of the royal mount that went through a colossal gold nugget was nothing more than an advertisement aimed at the great North African merchants. The king was richly adorned with silk, had standards carried during his official outings, and marched behind giraffes, elephants, and “other kinds of wild animals.” The sequence was intended to demonstrate the domestication of the stark, brutal, and uncivilized African countryside so feared by northerners.

This king—we don’t know his name—was Muslim. That was not the case a century earlier (7). Following conversion, the African kingdom consolidated the dominant position that it had already acquired toward the middle of the preceding century, when al-Bakrî described in detail the pomp of a king of a traditional religion, “pagan” according to the Andalusian author. We don’t know when or through what process this conversion and the ascent to power of a Muslim dynasty (if the king was not of the same line as his predecessor) occurred. Nor why—perhaps as a result of troubles caused by the rise of the Almoravids, that juridical and censorious movement, which was long thought to have razed Ghâna to the ground and converted its inhabitants by force. Or perhaps it was a result of the exploitation of new gold deposits, those of Bouré, in present-day Mali. But whatever the causes, we know the consequences: the dynasty, like all dynasties, chose for itself a new genealogy, whether real or imaginary. The sovereign of a few decades prior still had at his service priests tasked with maintaining the cult of the ancestors; the present one, in contrast, chose Alî ibn Abî Tâlib, cousin and son-in-law of the Prophet, for a distant ancestor and protector. The king of Ghâna’s new adherence to the Muslim faith, the affiliation with the community of believers, was a guarantee for the merchants, a guarantee of the application, in these distant lands, of a commercial law that was deemed equitable. As a result, was not the sultan reputed to be just? Twice a day, the king rode at the head of his officers “through the lanes of the town and around it. Anyone who has suffered injustice or misfortune confronts him, and stays there until the wrong is remedied.”

Al-Idrîsî’s description allows us to measure the religious and institutional evolution that the kingdom underwent during the decades since al-Bakrî. But we have a more serious problem: eleventh-century Ghâna was located in the Aoukar, southernmost Mauritania, where traces of it are even believed to have been found. Twelfth-century Ghâna, according to al-Idrîsî, was next to a river, and straddled its banks. Several details strengthen this version: the inhabitants “have strongly made boats . . . which they use for fishing or for moving about between the two towns,” and the sultan’s palace itself was built right next to the river.

On the pretext that al-Idrîsî was often unreliable, this part of his description has sometimes been dismissed, while the portion describing the kingdom’s conversion, or at least that of its sovereign, to Islam has been retained. Archaeological traces of the city described by al-Bakrî would be sought after; but we have never really looked for al-Idrîsî’s. Yet there are good reasons to think that it existed: for, however odd it may have sounded to him, al-Idrîsî clearly states that it was made up of two cities, situated on the two banks of a river; like that of the Aoukar it consisted of two principal entities, the commercial city and the royal city. Here the river would have promoted and maintained the capital’s spatial and cultural bipolarity, all while drawing the communities set up on its banks physically closer together. Then we know the date of the construction of the palace near the river: it was recent (in al-Idrîsî’s day). By itself this clue pleads for a new foundation, a royal act; in other words, for a relocation of the capital, as a new city would require a new palace. Let’s call the city of al-Bakrî Ghâna I and al-Idrîsî’s Ghâna II. All things considered, Ghâna II was only a few days to the south of Ghâna I, a shift from the steppe to the savanna, which would have carried the name of the kingdom with it. Nevertheless, this displacement raises a series of questions: on the bank of what river—the Senegal or the Niger—and in which areas that archaeologists have not surveyed was Ghâna located? And what became of the first? By what name was our Ghâna I then known if the capital was moved to Ghâna II? And, more importantly, what exactly was transferred from one capital to the other: a royal seat and its officials? A whole population? Or is it possible that this political transfer reflects not a physical movement, but the seizure of power, and the subsequent capture of the title of the sovereign of Ghâna, by a southern vassal from a weakened overlord? For al-Idrîsî and the Maghrebian merchants only one thing was useful to know: the greatest market of the Land of the Blacks could still be found in the shadow of the royal palace.

BIBLIOGRAPHICAL NOTE

The extracts from the text of al-Idrîsî are taken from Nehemia Levtzion and J.F.P. Hopkins (eds.), Corpus of Early Arabic Sources for West African History (Princeton, NJ: Markus Wiener, 2000), pp. 109–110; further bibliography on al-Idrîsî is provided in chapter 14. Nehemia Levtzion, “Ancient Ghana: A Reassessment of Some Arabic Sources,” in Le Sol, la parole et l’écrit: 2000 ans d’histoire africaine. Mélanges en hommage à Raymond Mauny (Paris: Société française d’histoire d’outre-mer, 1981), 1:429–437, put forth the hypothesis that the transfer of the capital could have been a consequence of the Almoravid raid of 1076. However, the reality of this raid, beginning with the destruction of the city of Ghâna, has been called into question by David Conrad and Humphrey Fisher, “The Conquest That Never Was: Ghana and the Almoravids, 1076. I. The External Arabic Sources,” History in Africa 9 (1982): 21–59, and “The Conquest That Never Was: Ghana and the Almoravids, 1076. II. The Local Oral Sources,” History in Africa 10 (1983): 53–78. The critical opinion put forward in these articles is nowadays generally accepted, but Nehemia Levtzion, undoubtedly one of the premier scholars on the subject, and certainly not an uncritical reader of sources, made clear that he remained unconvinced. See his “Berber Nomads and Sudanese States: The Historiography of the Desert-Sahel Interface,” in his collection of varia: Islam in West Africa: Religion, Society and Politics to 1800 (London: Variorum, 2004), item X.