Mali Empire

A song for a founding

formationDrafted from Gomez, African Dominion: A New History of Empire in Early and Medieval West Africa (2018)

Almost everything known about the founding of Mali comes from a poem that was performed for seven centuries before anybody wrote it down.

The Sundiata epic is the property of the jeliw — griots, hereditary specialists in genealogy, praise and history, whose families have been attached to the same lineages for generations. It tells of a crippled boy, son of a Malinke king and a hunchbacked woman, exiled by a rival queen, who returns to overthrow Sumanguru of Sosso — a sorcerer-king — at a battle called Kirina, and founds an empire.

It was first transcribed in the twentieth century.

What is and is not in doubt

The event is not seriously doubted. There was a Sosso power, Mali replaced it, and the Arabic sources writing a century later know Sundiata's name — Ibn Khaldun has him as Mari Jata, founder of the dynasty. What is uncertain is the date, which is conventional; the sequence; and how much of the narrative is thirteenth century and how much is the accumulated shaping of a text whose whole purpose is to be recited well.

Gomez treats the oral tradition as evidence to be handled carefully rather than discarded, which was not the default when the older narratives of West Africa were written, and it changes a great deal about what can be said.

What a griot is for

It is worth being precise, because "oral tradition" invites a reader to imagine folklore.

The jeliw are a caste. They marry within it, they hold hereditary attachments to particular noble families, and their function is professional: to know the genealogy, to recite it accurately, to praise, and on occasion to speak to a ruler in terms nobody else may use. A dynasty's legitimacy is partly held in the memory of specialists who are paid to hold it.

The Keita descent from Sundiata is a claim in the ordinary sense — publicly made, and the ground of a right to rule — and the institution that maintains it is a profession rather than a chancery.

There is a second descent claim beside it and it points the other way entirely. In the Arabic sources the Keita trace themselves to Bilal ibn Rabah, the Abyssinian whom the Prophet made the first muezzin. That makes the dynasty's Islam ancestral rather than adopted, and it is aimed at an audience in Cairo and Fez rather than on the Niger.

A dynasty maintaining two genealogies for two audiences is not unusual — the kings of Pontus claimed Darius and Alexander at the same time, for the same reason. What is unusual here is that one of Mali's was kept by a hereditary profession, in performance, and the other was written down in a language most of the empire could not read.

The shape of the thing

The empire that followed Kirina was not a territorial administration in the way a Mediterranean empire was.

It was a core on the upper Niger, a set of provinces under appointed governors, and beyond them a wide zone of tributaries, allied chiefs and trading towns that paid, sent contingents, and otherwise ran themselves. Al-Umari, writing in Cairo from informants, describes fourteen provinces. Ibn Battuta, who was actually there in 1352, describes a court with elaborate protocol, a functioning judiciary and roads safe enough for a traveller to use alone — which he says explicitly and which he did not say about most places he went.

The gold they did not own

expansionDrafted from Levtzion, Ancient Ghana and Mali (1973)

Mali did not own the goldfields.

The gold came from Bambuk, between the Senegal and the Falémé, and from Bure on the upper Niger, and later from the Akan forest further south. The people who worked it were not Malian subjects in the ordinary sense, and were not Muslim, and the empire's authority over them was a matter of arrangement rather than administration.

The story everybody tells about this comes from al-Umari, writing in Cairo in the 1330s from people who had travelled with Mansa Musa. A mansa attempted to bring the gold-producing country under direct rule and to convert its people. The producers responded by not producing. The arrangement was restored, and left alone thereafter.

Whether it happened as told matters less than the fact that this was the account the Malians themselves gave of why their empire stopped at a particular line on the map.

What the state took instead

A royalty on production, and a tax on movement.

All gold nuggets belonged to the mansa; only gold dust circulated. That is a claim on the product rather than on the mine, and it is enforceable where the gold enters the empire rather than where it comes out of the ground. Beside it were the tolls on the trans-Saharan trade: salt coming south from Taghaza, gold going north, and copper, cloth, horses and books coming back.

A reader who assumes Mali owned the mines has the empire's whole constitution wrong. This was a state that taxed a commodity produced by people it had deliberately decided not to govern.

Salt

The other half of the trade gets less attention and was worth more per journey.

Salt came from Taghaza in the middle of the Sahara, where the houses and the mosque were reportedly built of salt blocks because there was nothing else to build with, and it went south in slabs. In the savanna it was scarce and necessary, and the exchange rate against gold at the southern end was close enough to parity in some accounts to have generated a good deal of European legend.

Levtzion, Ancient Ghana and Mali (1973) · Robin Taylor · CC BY 2.0

Mali sat between a region with gold and no salt and a region with salt and no gold, and taxed the traffic in both directions.

And the third commodity

Slaves, moving north.

The trans-Saharan slave trade ran for a millennium before the Atlantic one and continued alongside it, and the Niger empires were both suppliers and consumers. Mali's own army rested on it: free troops came through intermediaries who could withhold them, and a ruler who wanted a force that could not be withheld acquired one by buying and capturing people.

Put the two together — the royalty on gold and the purchased soldiery — and the shape of the state is clear enough. Its power came from controlling the movement of things and of people, not from administering territory.

The other end of the gold

peakDrafted from Levtzion, Ancient Ghana and Mali (1973)

Almost every state around the Mediterranean in this period struck gold coin, and a great deal of the gold came from West Africa, up the trans-Saharan routes through Sijilmasa and Awdaghust. The Almoravids were a Saharan movement sitting on the northern end of those routes. Mali held the southern end.

Sundiata Keita assembled it in the 1230s out of the Malinke chieftaincies of the upper Niger, on the ruins of Ghana. It controlled the goldfields of Bambuk and Bure, the Niger as a transport artery, and the caravan termini at Timbuktu and Gao. Mansa Musa's pilgrimage to Mecca in 1324 — with an entourage and quantity of gold that reportedly disturbed the Cairo market for years afterwards — is the event that put a West African ruler on European maps, and the Catalan Atlas of 1375 draws him holding a gold nugget.

Bibliothèque nationale de France · Levtzion, Ancient Ghana and Mali (1973) · Abraham Cresques (attributed) · Public domain

Mali at its widest, in 1380, is put at about 1,100,000 square kilometres — one of the larger states on earth at the time, and larger than anything in Latin Europe.

That figure should be read with a caution the empire itself demands. Its authority was strong along the river and the caravan routes and progressively weaker away from them, and what it actually controlled was a network rather than a territory. The area is the box the network sits in.

What it did propagate outward was Islam, and the distinction matters. The merchants brought it and the ruling class adopted it generations before Mansa Musa; what Mali did afterwards was endow institutions — Timbuktu and Djenné became centres of learning drawing scholars from the whole Maghreb — which is a different act from converting.

The pilgrimage, and what it cost

peakDrafted from Levtzion, Ancient Ghana and Mali (1973)

Mansa Musa left for Mecca in 1324 with an entourage the Egyptian sources describe in figures that should be treated cautiously, and gold in quantity. He spent and gave it away in Cairo on a scale that the sources say depressed the value of gold there for years afterwards.

Two things follow, and only one of them is usually told.

The first is reputation. Cairo was the informational hub of the Islamic world, and after 1324 West Africa was on the map — literally: the Catalan Atlas of 1375 draws a crowned Black king holding a gold nugget on the southern edge of the world, labelled as the lord of Mali. It is the first substantial European image of a sub-Saharan state.

The second is that the pilgrimage was a fiscal event for somebody else's economy. A single man's travelling expenses moved the price of gold in Cairo for years, which is a measure of how much of it there was and where it came from: Mali and the Maghreb were the two ends of one system, and the Mediterranean's bullion supply ran through it.

Musa returned with scholars and an architect, endowed mosques at Timbuktu and Gao, and made the pilgrimage route a channel for Maliki learning into the Niger bend. That is the part that mattered and it is not the conversion. The ruling class had been Muslim for generations. What the pilgrimage produced was institutions — mosques with endowments, schools with salaried teachers, a route along which scholars came south and Malian students went north — and institutions are what carry a religion once a court has adopted one.

Levtzion, Ancient Ghana and Mali (1973) · upyernoz · CC BY 2.0

Three crusts of bread

peakDrafted from Battuta, The Travels of Ibn Batuta (1829)

Ibn Battuta reached the Malian court in 1352, stayed eight months, disliked a good deal of it, and wrote the most detailed account anybody produced of a West African empire at its height. The dislike is what makes it useful.

He arrived sick and was well treated. He lodged with the resident foreigners; the qadi, a noted pilgrim, made him his guest and sent him a present and a cow; he was ill for two months.

Then Mansa Sulayman gave a feast, and Ibn Battuta was presented, and afterwards the sultan sent a gift to the judge's house. The qadi came hurrying to fetch him: Up, for the Sultan has sent you a present.

I hastened, expecting that a dress of honour, some horses, and other valuables, had been sent; but, behold! they were only three crusts of bread, with a piece of fried fish, and a dish of sour milk. I smiled at their simplicity, and the great value they set on such trifles as these.

Two months later he had still received nothing, so he stood up in the sultan's presence and said so:

I have travelled the world over, and have seen its kings; and now, I have been four months in thy territories, but no present, or even provision from thee, has yet reached me. Now, what shall I say of thee, when I shall be interrogated on the subject hereafter?

He got a house and provisions, and on leaving he got a hundred mithqals of gold. His summary judgement of the man who gave it to him is a most avaricious and worthless man.

The prostration

What he describes of the court protocol is precise and hostile in equal measure.

Of all people, he says, the blacks debase themselves most in the presence of their king. A man summoned before the sultan puts off his usual clothing and puts on a worn-out dress; he prostrates himself; and when the sultan addresses somebody, the man takes up dust and puts it on his head and shoulders. While the sultan speaks, everybody present stands with his turban off.

That reads as humiliation to a fourteenth-century Moroccan lawyer accustomed to the etiquette of Cairo and Delhi. It is a court ceremonial of extreme formality serving the same function that kneeling and uncovering served in Europe, and the useful thing about the passage is that he could not stop himself from writing down exactly what he saw before delivering his opinion of it.

What he admired

Two things, and both are worth having from a man who had been almost everywhere.

One of the best things in these parts is, the regard they pay to justice, for, in this respect, the Sultan regards neither little nor much. The safety, too, is very great; so that a traveller may proceed alone among them, without the least fear of a thief or robber.

Ibn Battuta had crossed the Delhi sultanate, Ilkhanid Persia, Mamluk Egypt and the roads of Yuan China. When he says the security of the Malian roads is the best he has met, that is a comparison with most of the fourteenth-century world.

He also admired the observance of prayer, and the insistence that children commit the Quran to memory — he describes seeing boys in chains until they had learned their portion, which he approved of.

What he condemned

The women, mostly. Female servants and slaves went naked; free women did not clothe themselves until marriage; women, including the sultan's own daughters, came into the royal presence unclothed. Some of the population still ate carrion and dogs and asses.

And the inheritance. Among the merchants at Walata further north he had already found that nobody was named after his father but after his maternal uncle, and that a man's heir was his sister's son and not his own — a custom I witnessed nowhere else, except among the infidel Hindoos of Malabar. What offended him was not the practice but the combination: these were Muslims who knew the Quran by heart and had an explicit inheritance law available to them, and were not using it.

Why the hostility is the evidence

A sympathetic visitor would have passed over all of it.

Almost everything specific that is known about how fourteenth-century Mali actually worked — the ceremonial, the inheritance, the position of women, the policing of the roads — survives because a foreign lawyer found it objectionable and could not resist recording precisely what he objected to. He is the only outsider who went, and the ways in which he was wrong about the place are the reason there is anything to be right about.

A nugget of gold sold cheap

contractionDrafted from (ed.), General History of Africa, Volume IV: Africa from the Twelfth to the Sixteenth Century (1984)

Ibn Battuta left Mali still under Mansa Sulayman, the king whose court he had found stingy and correct in equal measure. What followed Sulayman's own death, within a generation, turned an empire Ibn Battuta had described as safer to travel than almost anywhere else into one its own princes were actively looting.

A wife's conspiracy, and a son who did not last

Ibn Battuta's own account already hints at the trouble to come: he records that Sulayman's wife conspired against her husband during his stay, an internal quarrel the traveler treated as scandal rather than as the sign of institutional weakness it turned out to be. When Sulayman died, his son Fomba inherited the throne and held it barely a year before being deposed in 1359 by Mari Diata II, also known as Sundiata II - a name that borrowed the founder's own, and delivered nothing of his restraint.

An empire run into the ground

Mari Diata II's own reign is remembered by Mali's own later chroniclers as a catalogue of waste. He exhausted the royal treasury outright, and in the single act later writers treated as emblematic of the whole reign, sold off the dynasty's most famous treasure - a nugget of gold so large it was reckoned at twenty kintars in weight - to Egyptian merchants for a fraction of its worth. Whatever royal restraint had kept Mansa Musa's own fortune intact through a pilgrimage that gave away staggering quantities of gold in Cairo, Mari Diata II possessed none of it. He was eventually incapacitated by illness - sleeping sickness, by the account that survives - and removed from effective power before his own death.

A son on the throne, a general running the country

Mari Diata II's son, Mansa Musa III, held the throne from around 1374 to 1387, but the empire's actual governance during his reign belonged to his general, who restored a measure of order and put down a rebellion at the copper-trading town of Takedda. It was a pattern that would recur for the rest of the century: princesses and court factions intriguing over the succession, provincial governors increasingly ignoring instructions from Niani, and an imperial prestige that, however diminished in practice, still carried enough weight abroad that its actual unraveling was not yet visible from outside. That unraveling would become impossible to miss once Timbuktu itself, the empire's own greatest trading city, fell to raiders from the desert its own frontier had stopped being able to hold.

The long recession

contractionDrafted from Levtzion, Ancient Ghana and Mali (1973)

Mali has no conqueror and no date anybody would defend closely. What it has is a contraction lasting two and a half centuries.

What went, in order

Timbuktu in 1433, to Tuareg out of the desert. It was the first substantial loss, and it went to a power with no state behind it — which is the part worth noticing. The empire's frontier at the desert edge had stopped being defensible before any rival empire existed to test it.

Then Gao. Songhai had been a Malian province in the fourteenth century; by the 1460s it was taking the Niger bend for itself.

Then the Atlantic. Portuguese ships reached the Senegal in the 1440s and the Gold Coast in the 1470s, and gold that had gone north across the Sahara for six centuries began going west to the sea instead. Mali understood the problem directly: an embassy went to Portugal in 1534 asking for help against Songhai — the earliest recorded diplomatic contact between a West African empire and a European state — and got nothing.

The Atlantic is the one that mattered most and the one nobody in Mali could have done anything about. For six hundred years the gold of the western Sudan had gone north across the desert, and every state on that route — Ghana, Mali, the Almoravids, the Marinids, the cities of the Maghreb — had lived off the fact that there was no other way to move it. Ships changed that, and they changed it from outside the system entirely. A caravan can be taxed, raided or diverted. A Portuguese caravel calling at a river mouth on the coast cannot be reached from the upper Niger at all.

Mali's embassy to Lisbon in 1534 is the measure of how well its rulers understood the position and how little they could do about it. They asked the people who had taken their trade to help them against the people who were taking their territory.

A village of six thousand

contractionDrafted from Africanus, The History and Description of Africa (1896)

Around 1510 a Granadan diplomat travelling the Sudan for the sultan of Fez passed through the kingdom whose ruler had, two centuries earlier, spent so much gold in Cairo that the price of it was still depressed a decade later. His entry on Mali is short and entirely matter-of-fact, and it is the clearest measure anywhere of what had happened.

What was there

In this kingdome there is a large and ample village containing to the number of sixe thousand or mo families, and called Melli, whereof the whole kingdome is so named. And here the king hath his place of residence.

A village. Leo uses the word without irony and without comparison; he is simply recording the size of the place, and six thousand families is not small. But the same book calls Timbuktu a town and Gao a great city, and the seat of the mansas gets the word for a settlement.

What he found there was prosperous. The region yields great abundance of corn, flesh and cotton; there are many artisans and merchants everywhere; the inhabitants are rich and have plenty of wares; and the king honourably entertaineth all strangers. There are great numbers of mosques, priests and professors — and then a detail a scholar would notice, because Leo was one:

which professours read their lectures onely in the temples, bicause they haue no colleges at all.

Teaching in the mosque because there is no madrasa is not poverty; it is how most of the Islamic world taught. But Leo had just come from Timbuktu, where there were colleges and a book trade, and he mentions the absence because the comparison was in front of him.

The sentence that matters

Leo gives the Malians the highest praise in his whole account of the Sudan: they excell all other Negros in witte, ciuilitie, and industry, and were the first of them to take up Islam.

Then, in the same breath, the end of the story:

At length Ischia subdued the prince of this region, and made him his tributarie, and so oppressed him with greeuous exactions, that he was scarce able to maintaine his family.

Askia Muhammad of Songhai had reduced the mansa to a tributary and taxed him to the point where he could not keep his own household.

What that is worth

It is worth being careful about the evidence. Leo is writing from memory in Italy years afterwards, for a European readership, and the identification of every place in his Sudan chapters has been argued over. He almost certainly did not see the Malian court himself; the passage reads like a summary of what he was told on the road.

But the shape of it is confirmed by everything else. Mali had lost the gold routes to Songhai, lost the Sahel towns, lost Timbuktu, and by the 1540s was sending embassies to Portugal asking for help against its neighbours. A century later the mansas were one chiefdom among several on the upper Niger.

The empire whose ruler's pilgrimage is the single best-known fact about medieval West Africa lasted, in its reduced form, longer than most of the states that had taken pieces off it. What Leo saw was not a ruin. It was a rich agricultural district with a market, a mosque, a court that received strangers properly, and a ruler who could not pay his tax.

Niani

endDrafted from Levtzion, Ancient Ghana and Mali (1973)

Mali has no conqueror and no date anybody would defend closely, and its ending is a series of humiliations rather than an event.

Niani was sacked by Songhai in 1545 and briefly held. Morocco destroyed Songhai in 1591, and Mali outlived its destroyer by decades without being able to take any advantage of it. The Bamana of Segu finished the empire in the seventeenth century.

An empire that outlived its own destroyer

The sequence at the end is worth setting out because it is genuinely strange.

Songhai took Mali's provinces, its trade and eventually its capital. Morocco destroyed Songhai in 1591, with firearms carried across the Sahara. Mali was still there — reduced, but still there — and could take no advantage of it whatever, because the thing that had made the western Sudan worth fighting over had already moved to the coast.

What finished it was neither of them. The Bamana of Segu, a people who had been inside the empire, took what was left in the seventeenth century, and by then there was not much to take.

The jeliw

afterlifeDrafted from Gomez, African Dominion: A New History of Empire in Early and Medieval West Africa (2018)

The empire ended in the seventeenth century and the poem did not.

The Sundiata epic went on being performed, by the same hereditary families, for another three hundred years after there was no Mali to perform it for — through the Bamana states, through the Muslim jihads of the eighteenth and nineteenth centuries, through French colonial rule, through independence, and it is being performed now.

That is a longer continuous run than the empire managed, and it is the reason anybody can say anything at all about how Mali was founded.

What a griot is for

Not entertainment, and not history in the sense a chronicler would recognise.

The jeliw are a hereditary occupational group, attached by descent to particular lineages, whose professional business is genealogy, praise and the public memory of who is descended from whom. A family has its griots; the griots have that family's past; and the arrangement is inherited on both sides.

That produces a very specific kind of record. It is maintained by people with a professional stake in its being impressive, performed to audiences who know it and would notice a change, and transmitted with enough formal structure that it does not simply drift into invention — while being under continuous pressure to serve the present, because the point of a genealogy is what it entitles somebody to now.

Everything strong and everything unreliable about the Sundiata tradition follows from that. It is not a document, and it is not folklore either.

What was written down, and when

The epic reached print in the twentieth century, and the versions differ, which is what one would expect of a performed text with hundreds of performers.

The best-known English text descends from a recitation by Djeli Mamoudou Kouyaté, of Guinea, published in the 1960s — a man from a griot family, giving a version his family had held, at the moment several West African countries were becoming independent and looking for a usable past. Others have been recorded since, and they are recognisably the same story with different emphases and different lengths.

So the founding of the empire is known from a text that was fixed on paper seven hundred years after the events, by the profession that had kept it, in the decade the region was being decolonised.

Why that is not a disqualification

It would have been treated as one for most of the twentieth century, and that assumption has been going out for about fifty years.

An oral tradition maintained by professionals with a formal structure and a critical audience is not the same thing as a rumour, and it is not obviously worse evidence than a chronicle written up at a court two hundred years after the fact by somebody with a patron. Both need handling. Only one of them was routinely dismissed.

Gomez's argument, and it is now the ordinary position, is that the material has to be worked rather than either accepted whole or thrown away — and when it is worked, a good deal of it lines up with what Ibn Khaldun heard in Cairo and what Ibn Battuta saw on the Niger.

And the other inheritance

The instrument, not the story.

Ghana had taxed the transit and taken a royalty on the metal without owning the mines. Mali did the same thing. Songhai did the same thing after Mali. Every one of the three collected at the desert edge and let the goldfields alone, because the goldfields belonged to people who were not their subjects and who would stop digging if pressed.

Three empires, six hundred years, one fiscal system, and the only thing that ever broke it was a ship.