Kanem-Bornu

Eight hundred years

Drafted from (eds), The History of Islam in Africa (2000)formation

The Sayfawa held the throne at Lake Chad, by the king list, from about 1075 to 1846. That is something close to eight hundred years under one dynasty.

For comparison: the Capetians held France for three hundred and forty years in the direct line. The house of Osman held the Ottoman throne for six hundred and twenty-three. The Solomonic dynasty of Ethiopia managed seven hundred and four. The Sayfawa outlast all of them, and almost nobody outside the field has heard of them.

The king list itself — the girgam — is the reason anybody can say so, and it is a rare kind of document to have from a tradition that was not writing: a dynastic chronicle maintained over centuries, giving each mai's name, his mother's name, and the length of his reign.

Kanem, then Bornu

The state began east and north of the lake, in Kanem. Al-Muhallabi describes a kingdom there in the tenth century whose ruler his people treated as a god and who ate in private. Al-Bakri and later Ibn Sa'id add to it.

Around 1380, after generations of pressure from the Bulala, the court was driven out of Kanem altogether and crossed to the west and south of the lake, into Bornu, where it built a new capital at Ngazargamu and stayed.

Kanem and Bornu are conventionally treated as two kingdoms and are better understood as one. The dynasty crossed the lake, the chronicle crossed with it, the trade road it lived on ran to both, and the claim its rulers made about themselves did not change. Nothing stopped; the country moved.

The road

What the kingdom sat on was the shortest crossing of the Sahara.

The Fezzan road ran from Lake Chad north through Bilma and the Kawar oases to Murzuq and on to Tripoli, and it is the least bad of the trans-Saharan routes — shorter than the western routes that Ghana, Mali and Songhai lived on, with reliable wells at intervals. Salt came south from Bilma. Natron from the lake shore went north. Horses came south. And people went north, in numbers that made this one of the two great suppliers of the trans-Saharan slave trade for most of a millennium.

Everything the kingdom collected came off that road or off the tributaries beside it.

A pedigree from Yemen

The Sayfawa took their name from Sayf ibn Dhi Yazan, a Himyarite noble of the sixth century.

He is a real figure. In the Arabic accounts Sayf goes to Constantinople for help against the Ethiopians occupying Yemen, is refused, goes to Ctesiphon instead, and comes back with a Persian force that takes the country.

A dynasty at Lake Chad claiming descent from a prince in Yemen — five thousand kilometres away and five centuries earlier — is not a genealogy anybody defends as history. It is a manoeuvre, and it is one that African dynasties made repeatedly: the sultans of Kilwa claimed a prince of Shiraz, the mansas of Mali claimed Bilal the Prophet's muezzin, the emperors of Ethiopia claimed Solomon.

What all four wanted was the same thing, which was to be legible to the world they traded with. Three chose an Islamic pedigree and one a biblical one, and in every case the point was that a merchant or an envoy arriving from outside should find a ruling house he already knew how to place.

Fifteen slaves for one horse

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

Around 1510 Leo Africanus wrote down the exchange rate that ran the kingdom of Bornu, and the arrangement it produced, and it is worth having in his words because no modern summary improves on it.

The king, he says, has three thousand horsemen in continual readiness and a huge number of footmen, and his subjects are so obedient that when he commands them they arm and follow wherever he pleases. They pay him no tribute but the tithes of their corn, and he has no revenues at all to maintain his estate but onely such spoiles as he getteth from his next enimies by often inuasions and assaults.

He was at war with a people beyond the desert who had once crossed it in force and wasted a great part of his kingdom. So he sent for the merchants of Barbary and told them to bring him horses.

For in this countrey they vse to exchange horses for slaues, and to giue fifteene, and sometime twentie slaues for one horse.

Waiting for payment

What Leo describes next is the part that turns a price into a system.

The merchants brought the horses. They could not be paid immediately, because the king did not have the slaves yet — he would have them when he came back from raiding, and he raided once a year, at one appointed season.

So the merchants waited. Often for three months, Leo says. Sometimes, when the king did not bring home enough captives to satisfy everybody, for a whole year. He maintained them at his own expense the entire time, which was cheaper than losing his suppliers. And Leo says he met traders there who had given up: they had extended credit for a full year and had stopped expecting to see it.

A state whose only revenue is what it takes by force, buying on credit the horses it needs to take it, from merchants who have to camp in the capital until next season's raid pays them.

Why the horses

Because the country south of Lake Chad is open, and cavalry works there, and almost nobody in it had any.

A mounted force operating against unmounted village populations does not need to be large and does not need to fight battles. Bornu's raiding zone was the non-Muslim country to the south and south-west, and the annual expedition against it was not a war in any sense that had an objective beyond the captives themselves.

The horses did not breed well at that latitude — the trypanosomiasis belt lies just south — so the stock had to be replaced continually from north Africa, which is what kept the circuit running instead of letting it close. Bornu could never accumulate enough horses to stop buying them.

What Leo got wrong

He should not be trusted on everything.

He describes the people of Bornu as embracing no religion at all — neither Christians, Muslims nor Jews — living after a brutish manner, with wives and children in common and no proper names, each one nicknamed for his height or his fatness. He attributes this to a merchant who had lived among them.

Bornu had been ruled by a Muslim dynasty for four hundred years by 1510, had endowed a hostel for its own students in Cairo, and would shortly be corresponding with the Ottoman sultan. Leo was describing peoples on the kingdom's southern raiding frontier and had been told they were the kingdom, or was repeating something he half remembered.

Which is the standing warning about him. Where he is reporting prices, distances and customs he watched, he is exact and nobody else is available. Where he is passing on what he was told about places he did not enter, he is worth about as much as the man who told him.

The girgam

Drafted from (eds), The History of Islam in Africa (2000)aside?What aside means

The girgam is the king list of Kanem-Bornu. It gives, for each mai in sequence: his name, his father's name, his mother's name, and the length of his reign.

It runs to something over sixty rulers. It was maintained in Arabic, added to as reigns ended, and it is why anybody can say that one dynasty held the throne for close to eight hundred years and can name the man at each end of it.

What a king list is worth

Very little on its own and a great deal in combination.

A list of names and reign-lengths does not say how a state worked. It does not state a succession rule, describe an institution, or date an event beyond a reign. What it does is establish a sequence — and a sequence is the thing most often missing.

It is worth appreciating how rare that is. The Greco-Bactrian and Indo-Greek kings of the second century BC are known almost entirely from coins, and their order has to be reconstructed from which types were overstruck on which, with the result that several kings appear to have been issuing at once and nobody is sure who followed whom. Kanem-Bornu has the opposite problem: a clean sequence of sixty rulers over eight centuries, and very little else.

The mother's name

The girgam's habit of recording each mai's mother is not decoration.

Royal women in Kanem-Bornu held formal office. The magira — usually the mai's mother — and the magara, the senior royal sister, had their own establishments, their own revenues and a recognised voice in the succession. Several are recorded acting as regents. A king list that names the mother in every entry is a document produced by a state that thought the maternal line was part of the identification.

It is worth resisting the obvious inference. Prominent queen mothers do not by themselves mean the throne descended through women: Ottoman valide sultans, Ethiopian empresses and Asante queen mothers all held formal power under three quite different succession systems, only one of which was matrilineal. What the girgam demonstrates is that the maternal line was part of how a Sayfawa king was identified. Whether it was part of who was eligible to be one, no source says.

What else was written here

More than the girgam, and less than survives.

Kanem sent students to Cairo from the thirteenth century and maintained a hostel there for its own travellers — a madrasa endowment in a foreign capital, which is the act of a state that expected a continuing traffic of its own scholars. It corresponded with the Almohads in the twelfth century and with the Ottomans in the sixteenth; some of those letters survive on the other side.

And under Idris Alooma it produced a narrative history of its own, written by the mai's own imam, which is among the very few contemporary accounts of a sub-Saharan state written from inside it.

Horses and people

Drafted from (eds), The History of Islam in Africa (2000)peak

Bornu's army was heavy cavalry. The horses were imported across the Sahara, the riders wore quilted cotton armour and mail, and the horses were armoured too — a description European visitors in the nineteenth century still found startling, because it looked to them like the fourteenth century arriving on the far side of a desert.

A horse cost slaves. Contemporary rates in the central Sudan put a good imported horse at somewhere between ten and twenty people.

The closed circuit

The arrangement is worth stating baldly, because it is the whole military and fiscal system of the kingdom in one sentence.

Slaves fought — there were slave regiments, and the musketeers of the sixteenth century were trained slaves. And slaves paid for the cavalry that captured them, because the annual raiding south into the non-Muslim country produced the people who were exchanged north for the horses that made the raiding possible.

That circuit ran for centuries and it is tighter here than anywhere else in West Africa, because the commodity being bought is the thing that does the buying. A kingdom in that position cannot stop raiding without losing its cavalry, and cannot lose its cavalry without being raided itself.

Idris Alooma

The reign that begins around 1564 is when the state acquires firearms and an administration in the same decades.

Turkish musketeers came through Tripoli. A corps of slave gunmen was trained by them. Brick fortifications were built. A standing camp replaced the seasonal levy. Islamic law was applied through appointed qadis in place of customary courts, and the mai made the pilgrimage.

And it was written up. Ahmad ibn Furtu, Idris Alooma's imam, produced narrative accounts of the Bornu and Kanem campaigns, which are among the very few contemporary histories of a sub-Saharan African state written from inside it.

The comparison that matters is two thousand kilometres west and almost exactly contemporary. Songhai fell in 1591 to a Moroccan force that crossed the Sahara with arquebuses; Bornu had been buying muskets through Tripoli since the 1570s. Two Sudanic empires, the same decade, the same technology arriving from the same direction — one of them bought it and one of them was destroyed by it.

What nobody wrote down

How the throne passed.

The Sayfawa held it within one family for eight centuries, which is the longest run any dynasty anywhere managed, and no source states the rule that made it work. The girgam records each mai's mother's name beside his own, and the queen mother and the senior royal sister held formal offices with revenues and a voice in the succession — but a prominent queen mother is compatible with a patrilineal throne, and several other African and Islamic states had both.

So the longest-reigning dynasty in the world is one whose succession rule cannot be recovered. That is not an accident of survival so much as a consequence of what the girgam was for: it was a list of who had reigned, kept by people who knew perfectly well how the next one would be chosen and had no reason to write it down.

Idris Alooma

Drafted from (eds), The History of Islam in Africa (2000)peak

The reign that begins around 1564 is the kingdom's high point, and what makes it one is that four separate things arrive together: firearms, a standing force, a legal administration, and a historian.

The guns

Turkish musketeers came south through Tripoli. Idris Alooma hired them, had them train a corps of slave gunmen, and used both alongside the mail-clad heavy cavalry the state had always run on.

The timing is worth setting beside what happened two thousand kilometres west. Songhai ended in 1591 at Tondibi, where a Moroccan force of a few thousand with firearms destroyed an army many times its size — about as clean an instance of gunpowder deciding an engagement outright as the century produced.

Idris Alooma began buying the same technology through the same desert twenty years before that battle. Two Sudanic empires, one generation, the same weapon arriving from the north: one purchased it and one was destroyed by it.

That is not a moral about foresight. Bornu had a shorter road to the Mediterranean and an existing diplomatic relationship with the Ottomans, and Songhai had neither. But the two cases together are about as close to a controlled comparison as this period offers.

The administration

Brick fortifications, which is unusual in the region and expensive. A standing camp in place of a seasonal assembly. And qadis — appointed judges applying Islamic law — in place of customary courts.

That last is the one with consequences. A ruler who replaces customary adjudication with appointed jurists is taking authority away from the lineage heads who held it and attaching it to men he pays. It is the same manoeuvre Askia Muhammad made in Songhai seventy years earlier, and the same one Ethiopia and the Ottomans each made in their own vocabulary.

Ahmad ibn Furtu

The mai's imam wrote the campaigns up. Two books survive, on the Bornu wars and the Kanem wars, and they are contemporary, internal, and specific.

That is rare enough in African history before the nineteenth century to be worth naming precisely. Mali is known from Arabic geographers who mostly had not been there, and from an epic written down in the twentieth century. Ghana is known from al-Bakri, at Córdoba, from what merchants told him. Songhai is known from two chronicles composed at Timbuktu a generation after the state fell, by men whose families its founder had persecuted. Great Zimbabwe has no text at all.

Ibn Furtu was there, worked for the man he was describing, and wrote while it was happening. That makes him partial in the ordinary way a court historian is partial in the ordinary way, and it makes this the one reign in eight centuries where the inside of the state is visible.

And what it did not change

The succession, and the shape of the economy.

Neither moved. The circuit — raiding south for people, exchanging people north for horses, using the horses to raid — went on running exactly as before, and the musketeers were bought with the same currency as the horses had been.

A state can acquire firearms, a judiciary and a court historian in a single reign and change nothing whatever about what it lives on. That is worth saying, because the reign reads in summary like a modernisation and it was not one. It was a better-equipped version of the same machine.

The letters

Drafted from (eds), The History of Islam in Africa (2000)contraction

In 1804 Usman dan Fodio declared jihad in Hausaland, west of Bornu, and within a few years his followers had overthrown the Hausa states and founded the Sokoto caliphate. In 1808 the movement took Ngazargamu, Bornu's capital, and the mai fled.

The state was recovered by a scholar from the north — Muhammad al-Amin al-Kanemi — who raised a force, drove the Fulani back, and then did something worth dwelling on. He wrote to them.

The argument

Al-Kanemi's objection was not military and not political. It was that the jihad was illegitimate, because Bornu was a Muslim state and Muslims may not be attacked as unbelievers.

The correspondence with Usman dan Fodio and later with his son Muhammad Bello survives on both sides, and it is a genuine theological exchange conducted while the two parties were at war. Al-Kanemi's case runs roughly: Bornu has been Muslim for seven hundred years, its rulers pray and fast and endow scholarship, and the practices you object to — surviving customary usages, laxity among the uneducated, rulers behaving badly — are found in every Muslim country including your own. If those make a country's people unbelievers, then no country is Muslim. Show me the ruling that makes them so.

The Sokoto side replied that the practices in question amounted to unbelief, that a ruler who tolerates them is complicit, and cited authorities.

Al-Kanemi answered the citations.

What the exchange actually settles

Nothing, in the end, and that is the interesting part.

The question at issue — when is a Muslim state Muslim enough that another Muslim state may not attack it — has no procedure for deciding it. There was no court either side could appeal to, no authority whose ruling would have bound them, and no possibility of a verdict. Two rulers argued it in writing, in good Arabic, citing the same authorities, and then fought about it.

What the letters do is make visible something that is usually invisible in religious wars, which is that both sides had a serious case and knew it. Sokoto's position was that a ruler who tolerates unbelief among his subjects is complicit in it. Al-Kanemi's was that unbelief has a definition, that his people met the tests, and that calling a Muslim an unbeliever in order to make him lawful prey is itself a grave offence in the law both sides accepted.

He was answering an accusation on which his country's existence depended, and he answered it by argument, and he was writing to men who were entirely capable of following the argument.

Who governed afterwards

Al-Kanemi saved the state and then held it. The mai went on reigning at Kukawa and al-Kanemi and then his son governed, for thirty-eight years, with the formal ruler in place and the actual one beside him.

Ethiopia had gone through the same thing a generation earlier and had a name for it — the Zemene Mesafint, the era of the princes, when the emperors at Gondar reigned without ruling while regional lords fought over which of them would hold the throne. Bornu's version had one lord instead of several and ended more decisively: in 1846 al-Kanemi's son killed the last mai and took the throne himself.

Eight hundred years of one dynasty ended in a room, and the man who ended it was the grandson of the scholar who had saved the state from the Fulani.

What 1846 was

Not a succession crisis, and not a conquest either.

It was a household officer's grandson deciding that the arrangement he had inherited — a reigning king and a governing minister — had become pointless, and removing the king. The Sayfawa had held the throne since about 1075. They lost it to the family of a scholar who had come to their rescue two generations earlier and had never gone home.

That is a remarkably common way for long dynasties to end and it almost never gets described as such, because it makes a poor story. Nobody invaded. The state carried on with the same borders, the same capital and the same administration. What changed was whose name was said.

A cage of cane

Drafted from Oudney, Narrative of Travels and Discoveries in Northern and Central Africa, in the Years 1822, 1823, and 1824 (1826)contraction

On 17 February 1823 a British officer riding ahead of his party came out of the trees a few miles short of Kuka and stopped, because there was a line of cavalry in front of him extending right and left as far as he could see.

Dixon Denham had spent the approach genuinely unsure what he was going to find. The reports had been contradictory — that the sheikh's soldiers were a few ragged men with spears living off raids, or that his forces were numerous and to a degree well trained — and he records the uncertainty in the sentence before he finds out:

we advanced towards the town of Kouka in a most interesting state of uncertainty, whether we should find its chief at the head of thousands, or be received by him under a tree, surrounded by a few naked slaves.

What the reception was for

Several thousand horse, standing quite steady, without noise or confusion. Then the welcome, which Denham understood perfectly well was a demonstration.

Three bodies detached from the centre and each flank and charged at the visitors, stopping a few feet from their horses' heads without checking speed until the moment of the halt, shaking spears and shouting Barca! barca! — blessing, blessing, sons of your country — then wheeling back to do it again. Meanwhile the wings closed in until the small Tripolitan party was entirely surrounded, packed so close as to be nearly smothered and in some danger from the crowding of the horses and the clashing of spears. Their chief was enraged and could do nothing; he was answered only by shrieks of welcome and spears rattled over his head.

I am quite sure this was premeditated.

The sheikh's own men — black chiefs and favourites, all raised to that rank by some act of bravery — wore mail of iron chain from throat to knee, split behind to hang down either side of the horse, some with iron skullcaps and chin-pieces, and their horses' heads plated as well.

The sultan behind the lattice

A fortnight later they were taken to Birnie to be presented to the actual king of Bornu, and Denham's account of that audience is worth giving in full.

The courtiers rode up, dismounted, prostrated themselves, and sat down on the ground in front of the ruler with their backs to him, as the custom required. The sultan was seated in a sort of cage of cane or wood by the door of his garden, and looked out at the assembly through the railing. Of him the visitors got only a glimpse through the lattice-work: his turban was larger than anybody else's, and his face from the nose downwards was completely covered.

The courtiers Denham describes at length and with open contempt. Large bellies and large heads being indispensable at that court, those not naturally equipped made up the deficiency with wadding, so that the belly appeared to hang over the pommel of the saddle; eight, ten and twelve shirts of different colours worn one over another helped; the head was wrapped in folds of muslin so as to deform it as much as possible, the most studied turbans having the effect of setting the head completely on one side. All of them hung about with charms in little red leather parcels, and the horses too.

Then the judgement:

Here was all the outward show of pomp and grandeur, without one particle of the staple commodity, power, to plead its excuse; he reigns and governs by the sufferance of the sheikh.

What Denham was looking at

Two things, and he only understood one of them.

The man in the cage was the mai of Bornu, of a family that had been reigning somewhere in the Chad basin since the eleventh century. The man who had put the cavalry on the road was Muhammad al-Kanemi, a scholar from Kanem who had raised the levies that saved the country from the Sokoto jihad fifteen years earlier, and had been running it ever since without taking the title.

Denham read the arrangement as an absurdity: pomp without power, a figurehead amused with ceremony so that the real ruler could govern more easily. That is accurate as far as it goes. What it misses is that the ceremony was the thing being preserved. A soldier who had saved the dynasty could not simply replace it without becoming another usurper in a region full of them, and the cage of cane was the price of legitimacy.

It held for twenty-three more years. In 1846 al-Kanemi's son had the last mai killed and took the title himself, and a line that had been on the throne since before the Norman conquest of England ended in a palace killing that nobody outside the Chad basin noticed.

1893

Drafted from (eds), The History of Islam in Africa (2000)end

The last century of it runs: the Fulani jihad taking Ngazargamu in 1808, al-Kanemi recovering the state and then governing it for thirty-eight years while the mai reigned, and the killing of the last mai in 1846 by al-Kanemi's son.

The end

The state continued under the Kanemi shehus for another forty-seven years, moved its capital to Kukawa, and was still functioning when Europeans began arriving overland to describe it — Denham and Clapperton in the 1820s, Barth in the 1850s, whose account is the fullest description of the kingdom by any outsider.

In 1893 Rabih az-Zubayr, a Sudanese warlord with a firearms-equipped army built out of the wreckage of the Egyptian slaving administration in the south, took the country and killed its ruler. The French killed Rabih in 1900, and the territory was partitioned between France, Britain and Germany.

What actually ended it

Not Europe, in the first instance. A slave raider.

Rabih az-Zubayr had been an officer of Zubayr Pasha's private empire in the Bahr al-Ghazal, the enormous slaving concern that the Egyptian administration in the Sudan had licensed and then tried to absorb. When that collapsed, Rabih went west with a core of trained riflemen and spent twenty years assembling a state out of raiding. By 1893 he had enough men and enough guns to take Bornu outright, and he burned Kukawa and killed its ruler.

Eight hundred years of dynasty, and then forty-seven of the family that replaced it, ended at the hands of a man who had begun as somebody else's enforcer in a business Bornu itself had been running for a millennium.

He held it for seven years. A French column killed him in 1900, and the territory was divided between France, Britain and Germany — which is why Kanem is in Chad, Bornu is in Nigeria, and the lake the kingdom was named for now has four countries on it.

Four countries on one lake

Drafted from (eds), The History of Islam in Africa (2000)afterlife

The kingdom's territory is now in four states and its lake is disappearing.

Rabih took Bornu in 1893 and was killed by a French column in 1900, and the partition that followed drew the boundaries where the three European powers' columns happened to meet. Kanem, where the dynasty began, is in Chad. Bornu, where it ended, is in north-eastern Nigeria. The rest is in Niger and Cameroon. Lake Chad, which the whole thing was organised around, has all four on its shore and has lost most of its surface area since the 1960s.

What the British did with it

They kept the dynasty, which is the standard colonial economy.

Indirect rule in northern Nigeria worked by finding an existing authority and governing through it, and Bornu had one — the Shehu, of al-Kanemi's line, the family that had displaced the Sayfawa in 1846. The British restored the office after Rabih's defeat, installed a shehu, and ran the emirate through him. The line is still there: the Shehu of Borno remains one of the senior traditional rulers in Nigeria, with a palace at Maiduguri and a ceremonial standing that long outlasted the colonial arrangement that preserved it.

So the succession runs: Sayfawa from about 1075 to 1846, al-Kanemi from 1846, interrupted by Rabih for seven years, restored under the British, and continuing. Somebody has been the ruler at Lake Chad, more or less continuously, for nine hundred years.

What the road was worth afterwards

Nothing, almost immediately.

The Fezzan road — Lake Chad to Bilma to Murzuq to Tripoli, the shortest crossing of the Sahara and the thing the kingdom had existed to tax — stopped mattering within a generation of the partition. The trans-Saharan caravan trade had already been undercut by steamships on the West African coast, and the traffic that had made a lake in the middle of the continent worth eight hundred years of dynastic attention simply moved to the sea.

A colonial boundary then cut the road in two, which finished it.

The girgam

What survives best is the list.

The girgam — sixty-odd rulers, each with his father's name, his mother's name and the length of his reign — was still being kept when Europeans arrived to read it, and it is the reason this kingdom can be discussed at all across the centuries when nobody outside was writing about it.

It is a modest document. It states no rule, describes no institution and explains nothing. What it does is establish that one family held a throne from the eleventh century to the nineteenth — which is the single most remarkable fact about the political history of Africa, and it is known because somebody kept adding a line each time a king died.